JP5345024B2 - Three-dimensional acoustic encoding device, three-dimensional acoustic decoding device, encoding program, and decoding program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a 3-dimensional sound encoding device, a 3-dimensional sound decoding device, an encoding program and a decoding program, which have original sound quality and spatial impression of sound, with a suppressed required bit rate. <P>SOLUTION: The device includes: a means 103 for calculating a basic signal conversion factor which allocates the signal of the 3-dimensional sound system to a signal of a virtual sound system, in which the number of channels of the system is down-mixed; a means 104 for calculating a basic signal corresponding to a channel signal of the virtual sound system; a means 105 for calculating an auxiliary signal conversion coefficient for calculating an auxiliary signal; a means 106 for calculating the number of auxiliary signals corresponding to a difference between the number of channels of the 3-dimensional sound system and the number of the basic signals, according to the auxiliary signal conversion coefficient; and a means 107 for encoding the basic signal and the auxiliary signal. The auxiliary signal is a prediction residual signal expressed by a difference of the channel signal of the 3-dimensional sound system and the prediction signal of the channel signal. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to an encoding device and an encoding program for encoding a signal of a three-dimensional acoustic system constituted by a large number of channels, and a decoding device and a decoding program for decoding an acoustic signal of a three-dimensional acoustic system.

  Conventionally, as a method for encoding an acoustic signal, a method based on MP3 is known for low bit rate encoding. These are parametric encoding methods including a signal obtained by downmixing an original signal into two channels and a parameter for restoring the original signal from the downmix signal.

  An AAC (advanced audio coding) method is known as a method of encoding while maintaining the quality of a CD (compact disc). In advanced BS broadcasting from 2011, an existing method using AAC coding will be standardized in order to encode and transmit 22.2 channel sound. Furthermore, a technique for improving the AAC system and making the setting of the absolute audible threshold variable is known (see, for example, Patent Document 1).

JP 2001-343997 A

  However, the method based on MP3 not only deteriorates the spatial impression of sound when 3D sound is downmixed to two channels, but also has the problem that the transmission rate can be compressed but the sound quality also deteriorates as a feature of the parametric method. was there.

  In addition, since the method based on the AAC method is based on the conventional two-dimensional audio encoding, when encoding three-dimensional sound such as 22.2 channel sound, a high compression rate can be realized. There was a problem that I could not.

  In order to solve the above problem, an object of the present invention is to provide a new three-dimensional audio encoding device capable of suppressing not only the original sound quality but also the spatial impression of the original sound and suppressing the required bit rate. The object is to provide a three-dimensional acoustic decoding device, an encoding program, and a decoding program.

In order to solve the above-described problem, a three-dimensional acoustic encoding device according to the present invention includes an acoustic signal input unit that inputs each channel signal of the first three-dimensional acoustic system, and each channel signal of the second three-dimensional acoustic system. A basic signal conversion coefficient calculation unit for calculating a basic signal conversion coefficient for deriving a basic signal to be, a basic signal calculation unit for calculating the basic signal according to the basic signal conversion coefficient, and the first signal from the basic signal. An auxiliary signal conversion coefficient calculation unit for calculating an auxiliary signal conversion coefficient for deriving an auxiliary signal for assisting restoration to each channel signal of the three-dimensional acoustic system, and an auxiliary signal for calculating the auxiliary signal according to the auxiliary signal conversion coefficient A calculation unit; and an encoding unit that encodes the basic signal and the auxiliary signal, wherein the auxiliary signal includes the channel signal of the first three-dimensional acoustic system and the channel. A predicted residual signal represented by a difference between the predicted signal and the predicted signal, and the predicted signal is a linear operation of the channel signal of the first three-dimensional acoustic system other than the channel signal and the auxiliary signal conversion coefficient It is calculated by these. Further, the basic signal calculation unit is configured to calculate a plurality of channel signals selected from the channels of the first three-dimensional acoustic system adjacent to the channels of the second three-dimensional acoustic system according to the basic signal conversion coefficient. The basic signal is calculated by down-mixing, and the auxiliary signal calculation unit configures the first independent equation to be solved on the decoding side with the same number of independent equations as the number of channels of the first three-dimensional sound system. The number of auxiliary signals corresponding to the difference between the number of channels of the three-dimensional acoustic system and the number of basic signals is calculated according to the auxiliary signal conversion coefficient .

In the three-dimensional audio encoding device according to the present invention, the sound intensity and sound pressure of the first three-dimensional sound system and the sound intensity and sound of the second three-dimensional sound system at a sound receiving point. The basic signal conversion coefficient is determined so that the pressure matches .

In the three-dimensional acoustic coding apparatus according to the present invention, the auxiliary signal transform coefficient calculating unit calculates the auxiliary signal transform coefficient so as to minimize the energy of the prediction residual signal.

  Further, in the three-dimensional acoustic encoding apparatus according to the present invention, the auxiliary signal calculating means calculates the prediction residual signal by the same number as the number of channels of the three-dimensional acoustic system, and the prediction residual signal is calculated from the calculated number. The number of prediction residual signals corresponding to the difference between the number of channels of the three-dimensional acoustic system and the number of basic signals is selected in the order of increasing energy, and the prediction residual signal is used as an auxiliary signal. To do.

  In the three-dimensional audio encoding device according to the present invention, the auxiliary signal transform coefficient and the auxiliary signal are updated at regular intervals.

  Further, in the three-dimensional acoustic encoding apparatus according to the present invention, the encoding means encodes each encoded bit allocation of the basic signal and the auxiliary signal by increasing an allocation amount allocated to the basic signal. It is characterized by.

  Furthermore, in order to solve the above problems, a three-dimensional acoustic decoding apparatus according to the present invention is a three-dimensional acoustic decoding apparatus that decodes each channel signal of a three-dimensional acoustic system composed of a large number of channels. Decoding means for decoding the basic signal and the auxiliary signal encoded by the encoding device, the basic signal and the auxiliary signal decoded by the decoding means, and the basic signal conversion coefficient and auxiliary signal conversion coefficient to the 3 Signal restoration means for restoring each channel signal of the two-dimensional acoustic system by solving the equation.

Furthermore, in order to solve the above-described problems, the present invention is also characterized as a program for causing a computer to function as the three-dimensional acoustic encoding device .

Furthermore, in order to solve the above-described problems, the present invention is also characterized as a program for causing a computer to function as the three-dimensional acoustic decoding device .

  According to the present invention, since the acoustic signal of the three-dimensional acoustic system is separated into the basic signal and the auxiliary signal, and the acoustic physical quantity at the sound receiving point can be reproduced only with the basic signal, the bit rate for the auxiliary signal is reduced. As a result, not only the original sound quality but also the spatial impression of the original sound can be maintained, and the required bit rate can be suppressed for transmission.

1 is a configuration diagram of a three-dimensional audio encoding device according to an embodiment of the present invention. It is a block diagram of the three-dimensional audio decoding apparatus of one Example by this invention. It is a figure which shows the speaker arrangement example of 22 channels except the LFE channel of the 22.2 channel sound signal which is a three-dimensional sound signal. It is a figure which shows the speaker arrangement example of the virtual acoustic system which downmixed the three-dimensional acoustic system of one Example by this invention. It is a flowchart which shows operation | movement of the three-dimensional acoustic encoding apparatus of one Example by this invention. It is a flowchart which shows operation | movement of the three-dimensional audio decoding apparatus of one Example by this invention. It is the schematic of the system using the three-dimensional acoustic encoding apparatus and three-dimensional acoustic decoding apparatus by this invention.

  Sound coming from a certain direction is represented by acoustic physical quantities such as sound pressure and particle velocity. Among these physical quantities, acoustic intensity is known as acoustic energy passing through a unit area and a vector quantity representing its directionality, and its effectiveness has been shown in fields such as sound source direction estimation. .

  If the number of channels can be reduced while maintaining the same sound intensity and sound pressure at the listening point (sound receiving point) compared to the case where sound is reproduced with a three-dimensional sound system, the number of channels is reduced. Deterioration of spatial impression of sound can be suppressed. However, even if the number of channels is reduced in this way, the spatial impression of the sound will be degraded if the playback sound is heard at a position that is different from the listening point set at the stage of reducing the number of channels. . Therefore, when transmitting sound signals of a three-dimensional sound system with high quality, a signal (hereinafter referred to as “basic signal”) converted to a small number of channels while keeping the sound physical quantity such as sound intensity constant at a specific listening point. In addition, a signal for reproducing a spatial impression during reproduction (hereinafter referred to as “auxiliary signal”) is required.

  As described above, the present invention is a system characterized in that the sound signal of the three-dimensional sound system is transmitted by being divided into a basic signal and an auxiliary signal. Also, if the sound signal of the three-dimensional sound system can be divided into the basic signal and the auxiliary signal as described above, when encoding (quantization), bits are assigned to the basic signal with emphasis, and By reducing the allocation, it is possible to compress the bit rate while maintaining the spatial impression of the original sound.

  Hereinafter, the present invention will be described in detail with reference to the drawings. First, the three-dimensional acoustic encoding device and the three-dimensional acoustic decoding device of the present embodiment will be described.

[Configuration of 3D Acoustic Encoding Device and 3D Acoustic Decoding Device]
FIG. 7 is a schematic diagram of a system using the three-dimensional acoustic encoding device and the three-dimensional acoustic decoding device of the present embodiment. This system includes a three-dimensional acoustic encoding device 10, a transmission device 11, a reception device 12, and a three-dimensional acoustic decoding device 13. The three-dimensional acoustic encoding device 10 performs an encoding process of the acoustic signal of the three-dimensional acoustic system and outputs the encoded signal to the transmission device 11. The transmission device 11 modulates a signal input from the three-dimensional acoustic encoding device 10 and transmits the modulated signal to the reception device 12. The receiving device 12 receives and demodulates the signal transmitted from the transmitting device 11 and outputs the demodulated signal to the three-dimensional acoustic decoding device 13. The three-dimensional acoustic decoding device 13 performs a decoding process on a signal input from the receiving device 12.

  FIG. 1 is a configuration diagram of a three-dimensional acoustic encoding device 10 of the present invention. The three-dimensional acoustic encoding apparatus 10 includes an acoustic signal input unit 101, a speaker position information storage unit 102, a basic signal conversion coefficient calculation unit 103, a basic signal calculation unit 104, an auxiliary signal conversion coefficient calculation unit 105, an auxiliary signal A signal calculation unit 106, an encoding unit 107, and a multiplexing unit 108 are provided.

  The acoustic signal input unit 101 performs A / D conversion on an acoustic signal from each speaker of the original three-dimensional acoustic system that is input, and converts the digital acoustic signal into a basic signal calculation unit 104 and an auxiliary signal conversion coefficient calculation unit 105. And output to the auxiliary signal calculation unit 106.

も予め記憶される。 The speaker position information storage unit 102 stores in advance position information ξ = (σ, θ, φ) in polar coordinates of each speaker of the three-dimensional sound system. Also, the position information in polar coordinates of each speaker of the three-dimensional sound system (hereinafter referred to as “virtual sound system”) when the number of channels of the three-dimensional sound system is downmixed to a smaller number of channels.

Are also stored in advance.

  The basic signal conversion coefficient calculation unit 103 calculates a basic signal conversion coefficient used for the basic signal based on the speaker position information stored in the speaker position information storage unit 102, and sends the basic signal conversion coefficient to the basic signal calculation unit 104 and the encoding unit 107. Output. As will be described later, the basic signal conversion coefficient is determined so that the sound intensity and sound pressure of the three-dimensional sound system and the virtual sound system are equal. The basic signal conversion coefficient is a constant calculated before transmission of the basic signal.

  The basic signal calculation unit 104 calculates a basic signal from the acoustic signal input from the acoustic signal input unit 101 and the basic signal conversion coefficient input from the basic signal conversion coefficient calculation unit 103 based on an arithmetic expression of the basic signal. And output to the encoding unit 107. The calculation formula of the basic signal will be described later.

  The auxiliary signal conversion coefficient calculation unit 105 calculates an auxiliary signal conversion coefficient used for the auxiliary signal and outputs it to the auxiliary signal calculation unit 106 and the encoding unit 107. The auxiliary signal conversion coefficient is determined so as to minimize the prediction residual, as will be described later. Since the prediction residual changes according to the change of the signal, the auxiliary signal conversion coefficient is a variable that is updated at regular time intervals.

  The auxiliary signal calculation unit 106 calculates an auxiliary signal from the acoustic signal input from the acoustic signal input unit 101 and the auxiliary signal conversion coefficient input from the auxiliary signal conversion coefficient calculation unit 105 based on the arithmetic expression of the auxiliary signal. And output to the encoding unit 107. The calculation formula of the auxiliary signal will be described later.

  The encoding unit 107 includes a basic signal conversion coefficient calculated by the basic signal conversion coefficient calculation unit 103, a basic signal calculated by the basic signal calculation unit 104, an auxiliary signal conversion coefficient calculated by the auxiliary signal conversion coefficient calculation unit 105, and an auxiliary signal The auxiliary signal calculated by the calculation unit 106 is quantized and encoded. When encoding (quantizing), a large number of bits are allocated to the basic signal and a small number of bits are allocated to the auxiliary signal. Note that, unlike the auxiliary signal transform coefficient, the basic signal transform coefficient is not updated when it is determined first, and does not need to be quantized because the amount of information is not large.

  The multiplexing unit 108 multiplexes the basic signal conversion coefficient, the basic signal, the auxiliary signal conversion coefficient, and the auxiliary signal encoded by the encoding unit 107 and outputs the multiplexed signal to the transmission device 11. It goes without saying that basic signal conversion coefficients can be transmitted in advance without multiplexing, or can be transmitted without multiplexing the basic signal conversion coefficients, basic signals, auxiliary signal conversion coefficients, and auxiliary signals.

  FIG. 2 is a configuration diagram of the three-dimensional acoustic decoding device 13 of the present invention. The three-dimensional acoustic decoding device 13 includes a demultiplexing unit 131, a decoding unit 132, and a signal restoration unit 133.

  The demultiplexing unit 131 demultiplexes the multiplexed signal input from the receiving device 12 and outputs the basic signal conversion coefficient, the basic signal, the auxiliary signal conversion coefficient, and the auxiliary signal to the decoding unit 132. Of course, if the signal input from the receiving device 12 is not multiplexed, the demultiplexing process is not necessary.

  The decoding unit 132 performs decoding processing on the basic signal transform coefficient, the basic signal, the auxiliary signal transform coefficient, and the auxiliary signal input from the demultiplexing unit 131, and outputs them to the signal restoration unit 133.

  The signal restoration unit 133 solves the simultaneous equations using the basic signal conversion coefficient, the basic signal, the auxiliary signal conversion coefficient, and the auxiliary signal input from the decoding unit 132, thereby generating the acoustic signal s (t) of the three-dimensional acoustic system. To restore.

[Calculation of basic signal conversion coefficient and basic signal]
Next, calculation of the basic signal conversion coefficient by the basic signal conversion coefficient calculation unit 103 and calculation of the basic signal by the basic signal calculation unit 104 will be described. In a space where the reverberation can be ignored, the Fourier transform p (r, ω) of the sound pressure on the sound receiving point r = (xyz) ^T by the sound source on the position ξ = (ξ _x ξ _y ξ _z ) ^T is If the signal (acoustic signal) is s (t), it is given by equation (1). Here, T represents transposition of a vector or a matrix.

ここに、ｓ(ω)はｓ(t)のフーリエ変換、ｋ＝２π／λは、位相定数、波長定数、波数などと呼ばれる係数、λは音波の波長、Ｇは音源から単位距離の点での音圧と音源入力との比を表す定数である。また、同じ点における音の粒子速度のフーリエ変換u(r,ω)は、式（２）で表される。

Where s (ω) is the Fourier transform of s (t), k = 2π / λ is a coefficient called phase constant, wavelength constant, wave number, λ is the wavelength of the sound wave, and G is the unit distance from the sound source. Is a constant representing the ratio between the sound pressure and the sound source input. Further, the Fourier transform u (r, ω) of the sound particle velocity at the same point is expressed by the equation (2).

このとき、受音点ｒにおける時間平均音響インテンシティ（以下、単に「音響インテンシティ」という）は、式（３）で表される。音響インテンシティは、単位面積を通過する音のパワーの１周期にわたる平均を表す量である。

At this time, the time-average sound intensity at the sound receiving point r (hereinafter simply referred to as “sound intensity”) is expressed by Expression (3). The sound intensity is a quantity representing an average over one period of the power of sound passing through a unit area.

  Now, polar coordinates with the sound receiving point r as the origin are used, and the coordinates of the sound source position ξ are (σ, θ, φ). Where σ is the distance from the origin, θ is the azimuth angle, and φ is the elevation angle. At this time, Expression (3) is expressed by Expression (4).

従って、３次元音響システムを構成するスピーカの中から、１つを選んで、この位置をξとすると、式（４）は、このスピーカに、対応するチャネルの音響信号ｓ(ｔ)を入力したときに、受音点で観測される音響インテンシティを表すこととなる。

Therefore, if one of the speakers constituting the three-dimensional acoustic system is selected and this position is ξ, Equation (4) inputs the acoustic signal s (t) of the corresponding channel to this speaker. Sometimes it represents the sound intensity observed at the receiving point.

スピーカの入力信号をｑ_j(ｔ)、j＝１，…，ｍとする。このとき、仮想音響システムのスピーカから３個のスピーカを選び、これらのスピーカの再生音によって、３次元音響システムのあるチャネルの音を再現する場合を考える。この場合の受音点における音響インテンシティは、式（５）で表される。 Next, a method for converting a signal of the three-dimensional sound system into a virtual sound system having a smaller number of channels while maintaining sound intensity at the sound receiving point will be described. The number of speakers of the virtual acoustic system is m, and the polar coordinates of the speaker installation positions are

Assume that the speaker input signal is q _j (t), j = 1,. At this time, a case is considered in which three speakers are selected from the speakers of the virtual acoustic system, and the sound of a channel in the three-dimensional acoustic system is reproduced by the reproduced sound of these speakers. The sound intensity at the sound receiving point in this case is expressed by Expression (5).

  Here, the speaker input signal q (t) of the virtual acoustic system is expressed by Expression (6) by the sound source signal (acoustic signal) s (t) and the basic signal conversion coefficient w of a channel of the three-dimensional acoustic system. Let's say.

すると、式（５）は、式（７）に変形される。

Then, Expression (5) is transformed into Expression (7).

  Therefore, if the sound intensity of the equations (4) and (7) is equal, the channel signal s (t) corresponding to the speaker is placed at the speaker position ξ of the three-dimensional sound system by three speakers. It is possible to localize the sound image. Conversion to a small number of channels can be performed by obtaining the basic signal conversion coefficient w based on such a principle. Now, assuming that the right sides of Equation (4) and Equation (7) are equal, Equation (8) is obtained.

この式は、ｗ₁，ｗ₂，ｗ₃に関する２次の連立方程式であり、このままでは解くことができないため、それぞれの第１、２成分を、第３成分で除すことにより、式（９）を得る。

This equation is a quadratic simultaneous equation relating to w ₁ , w ₂ , and w ₃ and cannot be solved as it is. Therefore, by dividing the first and second components by the third component, the equation (9 )

さらに、拘束条件として、式（１０）で表わされる受音点における音圧の一致を加える。

Further, as a constraint condition, a coincidence of sound pressures at the sound receiving point represented by the equation (10) is added.

式（１０）から、式（１１）の条件式が得られる。

From Expression (10), the conditional expression of Expression (11) is obtained.

式（９）及び式（１１）より、式（１２）で表される線形連立方程式を得る。

From the equations (9) and (11), a linear simultaneous equation represented by the equation (12) is obtained.

である。式（１２）の解は、式（１３）で表される。

It is. The solution of equation (12) is represented by equation (13).

式（１３）によって求められる基本信号変換係数ｗ_１〜ｗ_３を用いることにより、３次元音響システムの１つのチャネル信号を、仮想音響システムの３個のスピーカによって形成することができるようになる。

By using the basic signal conversion coefficients w _{1 to} w ₃ obtained by Expression (13), one channel signal of the three-dimensional sound system can be formed by three speakers of the virtual sound system.

  Next, a method for forming one channel signal of the three-dimensional sound system by four speakers of the virtual sound system will be described. When four speakers are used, the sound intensity at the sound receiving point is given by Expression (14) as in Expression (7).

式（１４）と式（４）が一致する条件から、式（９）と同様な式（１５）を得る。

Expression (15) similar to Expression (9) is obtained from the condition where Expression (14) and Expression (4) match.

これに、式（１６）の音圧一致条件を加えても、まだ拘束条件が不足している。

Even if the sound pressure matching condition of Expression (16) is added to this, the constraint condition is still insufficient.

  Therefore, a control parameter (panning weight) g for adjusting the gain of the fourth speaker is introduced. As a result, the following simultaneous equations (17) are obtained.

である。式（１７）の解は、式（１８）で与えられる。

It is. The solution of equation (17) is given by equation (18).

式（１８）によって求められる基本信号変換係数ｗ₁〜ｗ₄を用いることにより、３次元音響システムの１つのチャネル信号を、４個のスピーカによって形成することができるようになる。

By using the basic signal conversion coefficients w _{1 to} w ₄ obtained by Expression (18), one channel signal of the three-dimensional sound system can be formed by four speakers.

In this way, the channel signal of the original three-dimensional sound system can be assigned to the converted virtual sound system signal. The number of channels of the three-dimensional sound system is N, the signal of the three-dimensional sound system is s ₁ (t) to s _N (t), the number of channels of the virtual sound system, that is, the number of basic signal channels is M, and the three-dimensional sound system If the basic signal conversion coefficients when assigning the i-th channel to the j-th channel of the virtual acoustic system are w _{j, i} , j = 1,..., M, i = 1,. The basic signals b ₁ (t) to b _M (t) can be expressed as Expression (19).

[Calculation of auxiliary signal conversion coefficient and auxiliary signal]
Next, calculation of the auxiliary signal conversion coefficient by the auxiliary signal conversion coefficient calculation unit 105 and calculation of the auxiliary signal by the auxiliary signal calculation unit 106 will be described. In order to restore the original signal from the basic signal and the auxiliary signal, the auxiliary signal also needs to be expressed by a weighted sum of the original signals. Further, it is desirable that the auxiliary signal has a property that it does not affect the sound quality even if a small bit rate is assigned. Therefore, in the present invention, in order to satisfy these conditions, a prediction residual signal is used as an auxiliary signal.

このときの予測残差信号ｅ(ｎ)は、式（２１）で表される。

The prediction residual signal e (n) at this time is expressed by Expression (21).

この予測残差信号ｅ(ｎ)は、元のチャネル信号ｘ(ｎ)の重み付け和となっている。

This prediction residual signal e (n) is a weighted sum of the original channel signal x (n).

When the square error sum J of the auxiliary signal e (n) from the time t ₁ to the time t ₂ is defined by the equation (22), the auxiliary signal conversion coefficient a of the equation (21) is minimized so as to minimize the square error sum J. _{If k} 1, k = 1,..., N−1 are determined, the influence on the sound quality is reduced even if a small bit rate is assigned.

  Since Equation (22) is a quadratic function related to the auxiliary signal conversion coefficient, Equation (22) is minimized by Equation (23) in which Equation (22) is partially differentiated with respect to each conversion coefficient and set to 0. The coefficient to be obtained can be obtained.

式（２３）より、式（２４）が得られる。

From equation (23), equation (24) is obtained.

式（２４）は、式（２５）のように表現できる。

Expression (24) can be expressed as Expression (25).

よって、補助信号変換係数ａ_ｋ，ｋ＝１，・・・，Ｎ−１は、式（２６）により算出することができる。

Therefore, the auxiliary signal conversion coefficients a _k , k = 1,..., N−1 can be calculated by Expression (26).

  On the other hand, when Expression (22) regarding the square error sum J is modified, Expression (27) is obtained.

二乗誤差和Ｊの最小値Ｊ_minは、式（２４）を式（２７）に代入することにより、式（２８）で表される。

The minimum value J _min of the square error sum J is expressed by Expression (28) by substituting Expression (24) into Expression (27).

で与えられる。 If the number of channels of the original three-dimensional sound system is N and the number of channels of the basic signal is M, the number of channels of the auxiliary signal is (N−M) in order to restore the sound signal of the three-dimensional sound system. . On the other hand, the prediction residual signal e (n) represented by the equation (21) can be calculated as many as the number N of channels of the three-dimensional acoustic system. Now, e _i (n), i = 1,..., N are prediction residual signals for the i-channel signal of the original signal, and J _min ⁽ⁱ⁾ , i = 1, ..., N. At this time, (N−M) channels are selected from J _min ⁽ⁱ⁾ in ascending order, and the selected signal is used as an auxiliary signal. However, if the value of J _min ⁽ⁱ⁾ is extremely small and affects the accuracy at the time of decoding, such a prediction residual signal e _i (n) is not selected as an auxiliary signal and is suitable for decoding. The auxiliary signal can be selected within the specified range. If the channel number of the auxiliary signal is k ₁ ,..., K _NM , the auxiliary signal is

Given in.

In calculating the auxiliary signal e _ki (n), the extent to which the range from the time t ₁ to the time t _{2 in} the equation (22) is set is determined in consideration of the following points. If this range is widened, the real-time property is impaired, and the square error sum J generally increases. Therefore, it is possible to improve sound quality by narrowing this range, but it is necessary to frequently transmit auxiliary signal conversion coefficients.

The auxiliary signal and the auxiliary signal conversion coefficient are updated every fixed time p seconds (for example, 0.5 seconds), and the q-th auxiliary signal (that is, the auxiliary signal from (q-1) × p seconds to q × p seconds) is changed. e _j ^(q) (t), j = 1,..., N−M, and the auxiliary signal conversion coefficients when e _j ⁽ⁱ⁾ (t) are calculated are a _{j, i} ^(q) , j = 1,..., NM, i = 1,. At this time, the auxiliary signal conversion coefficient is calculated by solving Expression (26) from the signal of the original three-dimensional sound system from (q−1) × p seconds to q × p seconds. The auxiliary signal is calculated by Expression (21) and can be expressed as Expression (29).

[Reconstruction of acoustic signal]
Next, restoration of acoustic signals by the signal restoration unit 133 will be described. The sound signal of the three-dimensional sound system can be restored from the basic signal conversion coefficient, the basic signal, the auxiliary signal conversion coefficient, and the auxiliary signal. The basic signal represented by Expression (19) can be represented as Expression (30), and the basic signal represented by Expression (29) can be represented as Expression (31).

When the composite matrix W ^(q) is defined by Equation (32), the acoustic signal s (t) of the three-dimensional acoustic system can be restored based on Equation (33) every p seconds.

[Example when transmitting 22-channel signal]
Hereinafter, an embodiment will be described in which a 22-channel signal excluding the LFE channel of the 22.2 channel audio signal for Super Hi-Vision, which is a three-dimensional audio signal, is divided into a basic signal and an auxiliary signal. FIG. 3 is a diagram illustrating an example of arrangement of 22-channel speakers. Twenty-two white circles in the figure indicate 22 channel speaker positions, and the shaded circle in the center indicates a sound receiving point.

  FIG. 4 is a diagram showing the speaker position of the 22-channel acoustic signal and the speaker position when this is reproduced by eight speakers. The horizontal axis is the horizontal angle, and the vertical axis is the elevation angle. The horizontal angle indicates the angle in the counterclockwise direction with the right side as viewed from the sound receiving position as the origin. The white circles in the figure indicate the 22-channel speaker positions of the original three-dimensional sound system, the black circles indicate the 8-channel speaker positions of the virtual sound system after conversion, and the double circles indicate the positions where the speaker positions before and after conversion overlap. Using the basic signal conversion coefficient described above, each signal of 22 channels (ch1 to ch22) of the three-dimensional sound system can be approximately assigned to a plurality of signals of 8 channels (nch1 to nch8) of the virtual sound system. .

  FIG. 5 is a flowchart showing the operation of the three-dimensional acoustic encoding device 10. In step S501, the speaker position information storage unit 102 stores in advance position information in polar coordinates of each speaker of the three-dimensional sound system and position information in polar coordinates of each speaker in the virtual sound system.

  In step S <b> 502, basic signal conversion coefficient calculation section 103 calculates basic signal conversion coefficients based on the speaker position information stored in speaker position information storage section 102. The basic signal conversion coefficient is determined before inputting the acoustic signal. An arithmetic expression for calculating the basic signal is also determined before inputting the acoustic signal.

Table 1 shows an example of assigning the ch_i (i = 1,..., 22) signal of the three-dimensional acoustic system to the ch_j (j = 1,..., 8) signal of the virtual acoustic system. The left column of Table 1 shows each channel of the three-dimensional sound system. The center column of Table 1 is a channel of the virtual acoustic system adjacent to the channel of the three-dimensional acoustic system shown in the left column, and the channel signal of the three-dimensional acoustic system is assigned according to the basic signal conversion coefficient. The right column of Table 1 is a value of a basic signal conversion coefficient used when assigning each channel of the three-dimensional sound system to each channel of the virtual sound system. w _{j, i} represents a basic signal conversion coefficient when assigning ch_i to ch_j.

  In step S503, the acoustic signal input unit 101 performs A / D conversion on the acoustic signal from each speaker of the input three-dimensional acoustic system, and converts the digital acoustic signal into the basic signal calculation unit 104, the auxiliary signal conversion coefficient. It outputs to the calculation part 105 and the auxiliary signal calculation part 106.

In step S504, the basic signal calculation unit 104 calculates a basic signal from the digital audio signal obtained in step S503 and the basic signal conversion coefficient w _{j, i} calculated in step S501 based on the basic signal arithmetic expression. Is calculated. When the 22-channel signal of the original three-dimensional acoustic system is set as s ₁ (t) to s ₂₂ (t) and the basic signal is set as b ₁ (t) to b ₈ (t), Expression (34) is established.

In step S505, the auxiliary signal conversion coefficient calculation unit 105 calculates auxiliary signal conversion coefficients a _{j, i} ^(q) , j = 1,..., 14, i = 1,. , 21 is calculated.

である。 In step S506, the auxiliary signal calculation unit 106 calculates auxiliary signals e _j ^(q) (t), j = 1,..., 14 based on the equation (29). At this time, the auxiliary signal can be expressed in the form of equation (31). here,

It is.

In step S507, the encoding unit 107 calculates the basic signal conversion coefficient w _{j, i} calculated in step S502, the basic signal b (t) calculated in step S504, and the auxiliary signal conversion coefficient a _{j, i} calculated in step S505. ^{(q) The} auxiliary signal e ^(q) (t) calculated in step S506 is encoded. At the time of encoding, a large number of bits are allocated to the 8-channel basic signal b (t), and a small number of bits are allocated to the 14-channel auxiliary signal e ^(q) (t) to increase the compression rate.

In step S508, the multiplexing unit 108, the basic signal conversion coefficient w _{j, i} encoded in step S507, the basic signal b (t), the auxiliary signal conversion coefficient a _{j, i} ^(q) , and the auxiliary signal e ^{( q)} (t) is multiplexed and output to the transmitter 11.

FIG. 6 is a flowchart showing the operation of the three-dimensional acoustic decoding device 13. In step S501, the demultiplexing unit 131 demultiplexes the multiplexed signal input from the receiving device 12, and extracts the basic signal b (t) and the auxiliary signal e ^(q) (t).

In step S602, the decoding unit 132 performs a decoding process on the basic signal b (t) and the auxiliary signal e ^(q) (t) demultiplexed in step S601.

In step S603, the signal restoration unit 133 decodes the basic signal conversion coefficient w _{j, i} decoded in step S602, the basic signal b (t), the auxiliary signal conversion coefficient a _{j, i} ^(q) , and the auxiliary signal e ^{( q)} The sound signal s (t) of the three-dimensional sound system is calculated from (t) and restored. Here, when the composite matrix W ⁽ⁱ⁾ is defined by Expression (32), the acoustic signal s (t) of the three-dimensional acoustic system is expressed by Expression (33) every predetermined time p seconds (for example, 0.5 seconds). Restore based on.

  Here, in order to function as the three-dimensional acoustic encoding device 10, a computer can be suitably used. Such a computer includes an acoustic signal input unit 101, a speaker position information storage unit 102, and a basic signal conversion coefficient. A control unit for causing the calculation unit 103, the basic signal calculation unit 104, the auxiliary signal conversion coefficient calculation unit 105, the auxiliary signal calculation unit 106, the encoding unit 107, and the multiplexing unit 108 to function as a central processing unit This can be realized by a storage unit including a device (CPU) and at least one memory.

  Further, by causing such a computer to execute a predetermined program by the CPU, an acoustic signal input unit 101, a speaker position information storage unit 102, a basic signal conversion coefficient calculation unit 103, a basic signal calculation unit 104, The functions of the auxiliary signal conversion coefficient calculation unit 105, the auxiliary signal calculation unit 106, the encoding unit 107, and the multiplexing unit 108 can be realized. Furthermore, the acoustic signal input unit 101, the speaker position information storage unit 102, the basic signal conversion coefficient calculation unit 103, the basic signal calculation unit 104, the auxiliary signal conversion coefficient calculation unit 105, the auxiliary signal calculation unit 106, The program for realizing the functions of the combining unit 107 and the multiplexing unit 108 can be stored in a predetermined area of the storage unit (memory). Such a storage unit can be composed of a RAM or the like inside the computer, or can be composed of an external storage device (for example, a hard disk). In addition, such a program can be configured by a part of software (stored in a ROM or an external storage device) on an OS used by a computer as the three-dimensional acoustic encoding device 10.

  Furthermore, a program for causing a computer that functions as the three-dimensional acoustic encoding device 10 to function as means as each component of the present invention can be recorded on a computer-readable recording medium.

  Similarly, in order to function as the three-dimensional acoustic decoding device 13, a computer can be preferably used, and such a computer is used for causing the demultiplexing unit 131, the decoding unit 132, and the signal restoration unit 133 to function. The control unit can be realized by a central processing unit (CPU) and a storage unit including at least one memory.

  In addition, by causing such a computer to execute a predetermined program by the CPU, a program for realizing the functions of the demultiplexing unit 131, the decoding unit 132, and the signal restoration unit 133 is stored in the storage unit (memory ) In a predetermined area. Such a storage unit can be composed of a RAM or the like inside the computer, or can be composed of an external storage device (for example, a hard disk). Further, such a program can be constituted by a part of software (stored in a ROM or an external storage device) on an OS used by a computer as the three-dimensional sound decoding device 13.

  Furthermore, a program for causing a computer that functions as the three-dimensional acoustic decoding device 13 to function as means as each component of the present invention can be recorded on a computer-readable recording medium.

  Although the above embodiments have been described as representative examples, it will be apparent to those skilled in the art that many changes and substitutions can be made within the spirit and scope of the invention. Accordingly, the present invention should not be construed as being limited by the above-described embodiments, and various modifications and changes can be made without departing from the scope of the claims. For example, although the example in which the basic signal is calculated by performing the downmix shown in FIG. 3 has been shown, in the present invention, the number of channels and the channel position (speaker position) at the time of the downmix can be arbitrarily set. Further, in this embodiment, all channel signals are used when calculating the prediction residual signal, but it is also possible to calculate using only the signals of the target channel and the channels near the target channel.

  Thus, according to the present invention, the signal of the three-dimensional acoustic system constituted by a large number of channels can be downmixed to a smaller number of signals, which is useful for any application for encoding and decoding an acoustic signal. is there.

DESCRIPTION OF SYMBOLS 101 Acoustic signal input part 102 Speaker position information storage part 103 Basic signal conversion coefficient calculation part 104 Basic signal calculation part 105 Auxiliary signal conversion coefficient calculation part 106 Auxiliary signal calculation part 107 Coding part 108 Multiplexing part 131 Demultiplexing part 132 Decoding Part 133 Signal restoration part

Claims (10)

An acoustic signal input unit for inputting each channel signal of the first three-dimensional acoustic system;
A basic signal conversion coefficient calculation unit for calculating a basic signal conversion coefficient for deriving a basic signal to be each channel signal of the second three-dimensional acoustic system;
A basic signal calculation unit for calculating the basic signal according to the basic signal conversion coefficient;
An auxiliary signal conversion coefficient calculation unit for calculating an auxiliary signal conversion coefficient for deriving an auxiliary signal for assisting restoration from the basic signal to each channel signal of the first three-dimensional acoustic system;
An auxiliary signal calculation unit for calculating the auxiliary signal according to the auxiliary signal conversion coefficient;
An encoding unit for encoding the basic signal and the auxiliary signal;
Equipped with a,
The auxiliary signal is a prediction residual signal represented by a difference between a channel signal of the first three-dimensional sound system and a prediction signal of the channel signal;
The prediction signal is the stereophonic sound encoder calculated you wherein Rukoto by linear calculation between the channel signal and the auxiliary signal conversion coefficient of the other than the channel signal first three-dimensional sound system. The basic signal calculation unit downmixes a plurality of channel signals selected from each channel of the first three-dimensional sound system adjacent to each channel of the second three-dimensional sound system according to the basic signal conversion coefficient. To calculate the basic signal,
The auxiliary signal calculation unit is configured to form the same number of independent equations as the number of channels of the first three-dimensional sound system to be solved on the decoding side and the basic number of channels of the first three-dimensional sound system. The three-dimensional acoustic encoding device according to claim 1, wherein the number of auxiliary signals corresponding to a difference from the number of signals is calculated according to the auxiliary signal conversion coefficient. The basic signal conversion coefficient calculation unit matches the sound intensity and sound pressure of the first three-dimensional sound system with the sound intensity and sound pressure of the second three-dimensional sound system at a sound receiving point. and determines the basic signal conversion coefficients as three-dimensional acoustic coding apparatus according to claim 1 or 2. Said auxiliary signal conversion coefficient calculation unit, before and calculates the auxiliary signal conversion coefficient so the energy to minimize the Ki予 Hakazansa signals, according to any one of claims 1 3 3D acoustic encoding device.   The auxiliary signal calculation unit calculates the same number of prediction residual signals as the number of channels of the first three-dimensional acoustic system, and the first three-dimensional acoustics are calculated in ascending order of energy of the prediction residual signal. The three-dimensional according to claim 4, wherein a number of prediction residual signals corresponding to a difference between the number of channels of the system and the number of basic signals are selected, and the prediction residual signals are used as auxiliary signals. Acoustic encoding device.   The three-dimensional acoustic encoding device according to any one of claims 1 to 5, wherein the auxiliary signal conversion coefficient and the auxiliary signal are updated at regular intervals.   7. The encoding unit according to claim 1, wherein the encoding unit performs encoding by assigning a higher allocation amount to the basic signal for each encoding bit allocation of the basic signal and the auxiliary signal. 8. The three-dimensional acoustic encoding device according to item. A three-dimensional sound decoding apparatus for decoding each channel signal of a three-dimensional sound system composed of a number of channels,
A decoding unit for decoding the basic signal and the auxiliary signal encoded by the encoding device according to any one of claims 1 to 7;
Each channel signal of the first three-dimensional acoustic system is obtained from the basic signal and the auxiliary signal decoded by the decoding unit, and the basic signal conversion coefficient and the auxiliary signal conversion coefficient. A signal restoration unit that solves and restores the same number of independent equations,
A three-dimensional acoustic decoding apparatus comprising:   A program for causing a computer to function as the three-dimensional acoustic encoding device according to any one of claims 1 to 7.   A program for causing a computer to function as the three-dimensional acoustic decoding device according to claim 8.

Images