JPH10257600A - Multichannel signal processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable control so as to use a sound field over wide range by performing sound field correction through a stereoscopic sound field corresponding to the position of listener. SOLUTION: After filtering processing through a pair of left side and right side sound image localization filters 21a, 21b, 21c and 21d so as to respectively localize sound images on the left side and right side of listener, rear surround signals SL and SR are added through adders 21h and 21i, attenuators 21j and 21k to set a coefficient gfr corresponding to the position information of listener and adders 211 and 21m to two-channel stereo signals L and R. On the other hand, the rear surround signals SL and SR themselves are reproduced by speakers 3SL and 3SR arranged at rear positions almost laterally symmetric to the listener through attenuators 21h and 21o to set a coefficient gr corresponding to the position information of listener so that high presence can be provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a surround reproduction for processing and reproducing a multi-channel audio signal, and more particularly, to a multi-channel for creating a suitable multi-channel sound field in a listening room such as a general home. The present invention relates to a signal processing method and apparatus.

[0002]

2. Description of the Related Art As a system for reproducing a multi-channel audio signal, a 3-1 system in HDTV (front left L
ch, right Rch, center Cch, surround Sch), a 4ch matrix system (prologic) using Dolby surround, and a discrete system (AC-3). In particular, many American movies have surround tracks on which Dolby surround sound processing has been performed, and in movie theaters, decoding is performed, and signals are supplied to a plurality of speakers to be reproduced. When this is created as video software, a sound that has been subjected to surround processing is taken and is commercially available. As a result, many Dolby surround-processed video tape software, laser disks, or DVD disks have been sold.

[0003]

By the way, it is general to adjust the sound field in the multi-channel reproduction using a plurality of speakers as described above by using a level and a delay amount. There are various restrictions on the listening room where is created. In particular, when the listener approaches the front screen to clearly view the image as in the case of accompanying the image, the listener may inevitably approach the front speaker and lose the sound field balance with the rear speaker. Here, when the rear loudspeaker is far away, it becomes particularly noticeable, and the rear sound becomes low and is canceled out, resulting in a front-only stereo sound field. For example, as shown in the characteristic diagram of the distance-intensity level shown in FIG. 25, near the front, as can be seen from the intensity level on the vertical axis relating to hearing (this is inversely proportional to the square of the horizontal axis distance), almost only the front is dominant. become.
On the other hand, when the user moves away from the screen and approaches the rear speaker, the rear sound becomes dominant at the rear as shown in FIG.
The sound from behind may be abnormally loud and unpleasant.

Accordingly, it is necessary to adjust the sound field in accordance with the position of the listener. Various methods have been proposed for detecting the position of the listener. For example, Tokiko 1-611
As disclosed in Japanese Patent Application Publication No. 90-90, there is an apparatus that detects a listener's position by transmitting an optical signal and an acoustic signal from the listener's position, and receiving and calculating the acoustic signal on the acoustic device side. In addition, as disclosed in Japanese Patent Application Laid-Open No. 1-276900, by using a speaker as a part of a sensor (sound-producing body) for detecting the position of the listener, the position of the listener at an arbitrary position in the sound field space can be determined. There are things that can be detected. Further, as disclosed in Japanese Patent Application Laid-Open No. 2-184196, a method for detecting the position of a listener by obtaining data corresponding to the position of a remote control transmitter on the basis of data received from a plurality of remote control receiving units. is there. However, all of these need to be constant without adjusting the sound field balance between the left and right. However, since these detection positions are distances from the speakers, only the unbalance is adjusted. However, it is not possible to detect the position information before and after the listener and adjust the sound field using the three-dimensional sound field in a wide sound field range accordingly.

As described above, adjusting the sound field at an arbitrary position in multi-channel reproduction is generally performed by setting a level and providing a delay amount according to the distance (position). Although it is attempted to make adjustment so that it is easy to hear, as can be understood from the characteristic diagram shown in FIG.
In reality, it is not possible to adjust to be usable in a wide range by adjusting the number of times.

In view of the above, the present invention has been made in view of the above points, and has been made so that sound field correction using a three-dimensional sound field can be performed in accordance with the position of a listener so that the sound field can be used in a wide range. It is an object of the present invention to provide a multi-channel signal processing method and apparatus that can perform the processing.

[0007]

In order to achieve the above object, a multi-channel signal processing method and apparatus according to the present invention comprises a pair of speakers arranged at a front position substantially symmetrical to a listener and a rear position. A multi-channel signal processing method and apparatus configured to reproduce a multi-channel signal including a pair of left and right front stereo signals and a rear signal from a pair of speakers arranged in a head-related transfer function for the rear signal A filter processing step and means for outputting a pair of filtered left and right rear processed signals realizing a convolver in which a filter coefficient is set based on the left and right rear processed signals filtered by the filter processing step. Adding step and means for adding to the front stereo signal, and before adding to the front stereo signal A first level setting step and means for adjusting and setting the level of a pair of left and right rear processed signals, and a second level setting for adjusting and setting the level of the rear signal according to the adjusted and set level of the rear processed signal It has steps and means.

[0008] Further, a step and means for supplying an output signal of the adding step and the means to a pair of speakers arranged at the front position, and a rear signal having passed through the second level setting step and means is arranged at the rear position. And a step of supplying to the pair of speakers described above.

[0009] Further, the apparatus further comprises a position information supply step and means for supplying position information of the listener, wherein the first level setting step and means add to the level of the stereo signal according to the position information. The level of the rear processing signal is adjusted and set.

The position information supply step and means may supply position information based on a forward movement and / or a backward movement of the listener.

The multi-channel signal is an AC-3 type signal having a 5-channel configuration including a 2-channel front stereo signal, a 1-channel center channel signal, and a 2-channel rear signal. And a center addition step and a means for attenuating the center channel signal and adding the attenuated center channel signal to a pair of left and right front stereo signals.

Further, the multi-channel signal is a DV signal.
A wideband audio signal recorded on a D audio disk, further comprising a decoding step of decoding a signal recorded on the DVD audio disk before the filtering step and the adding step. It is.

The filter processing step and means are arranged so that the filter coefficients Hl and Hr are Hl = (SF−AK) / (S ² −A ² ) Hr = (SK−) for each channel of the pair of left and right rear signals. AF) / (S ² −A ² ) (where S is a transfer function from the pair of speakers to the ear on the same side of the listener, A is a transfer function from the pair of speakers to the ear on the opposite side of the listener, F Is the transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the position where the sound image is to be localized to the opposite ear of the listener.
, And adds the filtered output with the filter coefficient set to Hl and the filtered output with the different filter coefficient Hr of the other channel based on the set filtered output, respectively, and filters the pair of added outputs. The processed signal is output as a pair of left and right rear processed signals.

In the filtering step, the filter coefficient P is P = (F + K) / (S + A) (where S is a transfer function from a pair of speakers to the ear on the same side of the listener, and A is a pair of speakers) , The transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the position where the sound image is to be localized to the opposite ear of the listener. Transfer function up to)
And the first filter processing output obtained by filtering the sum signal of the pair of left and right rear signals, and the filter coefficient N is set to N = (F−K) / (S−A). A sum signal and a difference signal of the first and second filtered outputs based on a second filtered output obtained by filtering the difference signal of And output it as a rear processing signal.

A method and apparatus for processing a multi-channel signal according to another invention comprises a pair of left and right front speakers formed from a pair of speakers arranged at a front position substantially symmetric to a listener and a pair of speakers arranged at a rear position. A multi-channel signal processing method for reproducing a multi-channel signal including a stereo signal and a rear signal, the filter realizing a convolver in which a filter coefficient based on a head-related transfer function is set for the front stereo signal A filtering step and a means for outputting a pair of processed left and right front processed signals; an adding step and a means for adding the pair of left and right front processed signals filtered by the filtering step to the left and right rear signals; A first method for adjusting and setting the level of the pair of left and right front processing signals to be added to the rear signal And Bell setting step and means, and has a second level setting step and means for adjusting and setting the level of the front stereophonic signals in response to the adjusting and setting the level of the front processing signal.

[0016] Further, a step and means for supplying an output signal of the adding step and the means to a pair of speakers arranged at the rear position, and a front stereo signal having passed through the second level setting step and means is transferred to the front position. The method further includes a step and means for supplying the pair of speakers arranged.

Further, the apparatus further comprises a position information supply step and means for supplying position information of the listener, wherein the first level setting step and means add to the level of the rear signal according to the position information. It is characterized in that the level of the forward processing signal is adjusted and set.

Further, the position information supply step and the means supply position information based on a forward movement and / or a backward movement of the listener.

The multi-channel signal is a 5-channel AC-3 type signal including a front stereo signal of two channels, a center channel signal of one channel and a rear signal of two channels, and attenuates the center channel signal. And a center adding step and means for adding to the pair of left and right front stereo signals.

Further, the multi-channel signal is a DV signal.
And a decoding processing step and means for decoding a wideband audio signal recorded on the D audio disk and the signal recorded on the DVD audio disk.

Further, the filtering step and the means include: for each channel of a pair of left and right front stereo signals,
Filter coefficient Hl and Hr is Hl = (SF-AK) / (S 2 -A 2) Hr = (SK-AF) / (S 2 -A 2) ( however, S is the same side of the listener from a pair of speakers A is the transfer function from the pair of speakers to the ear on the other side of the listener, F is the transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the pair of speakers to the ear on the same side of the listener. Transfer function from the position to be localized to the ear on the other side of the listener)
, And adds the filtered output with the filter coefficient set to Hl and the filtered output with the different filter coefficient Hr of the other channel based on the set filtered output, respectively, and filters the pair of added outputs. The processed signal is output as a pair of processed left and right front stereo processed signals.

Further, the filtering step and the filtering means are such that the filter coefficient P is P = (F + K) / (S + A) (where S is a transfer function from a pair of speakers to the ear on the same side of the listener, and A is a pair Transfer function from the speaker to the ear on the other side of the listener, F is the transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the position where the sound image is to be localized to the opposite side of the listener. Transfer function to the ear)
And the first filter processing output obtained by filtering the sum signal of the pair of left and right front stereo signals and the filter coefficient N is set to N = (F−K) / (S−A). A sum signal and a difference signal of the first and second filtered outputs are obtained based on a second filtered output obtained by filtering the difference signal of the stereo signal, and the sum signal and the difference signal are filtered. The signal is output as a pair of left and right front stereo processing signals.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A multi-channel signal processing method and apparatus according to the present invention can be adjusted so that a sound field can be corrected over a wide range by performing a sound field correction using a three-dimensional sound field according to the position of a listener. And reproduces a multi-channel signal including a pair of left and right front stereo signals and a rear signal from a pair of speakers arranged at a front position substantially symmetrical to a listener and a pair of speakers arranged at a rear position. And a filter processing step and means for outputting a pair of filtered left and right rear processing signals for realizing a convolver in which a filter coefficient based on a head-related transfer function is set for the rear signal. An adding step and means for adding a pair of left and right rear processed signals filtered by the above to the left and right front stereo signals, A first level setting step and means for adjusting and setting the level of the pair of left and right rear processed signal to be added to the stereo signal, and adjustment of the rear processing signal
A second level setting step and means for adjusting and setting the level of the rear signal according to the set level.

<Embodiment 1> FIG. 1 is a block diagram showing an embodiment 1 of an apparatus for implementing the above-described multi-channel signal processing method of the present invention. In FIG. 1, reference numeral 1 denotes a decoder corresponding to the AC-3 system of Dolby Laboratories, which converts a digital signal stream into two-channel front stereo signals L and R, a one-channel center channel signal C,
And 5 of the rear surround signals SL and SR of the two channels
The signal is decoded and separated into a channel signal and a low frequency signal of 100 Hz or lower. Usually, the low-frequency signal is reproduced by a super woofer or distributed to each channel. However, in this embodiment, the low-frequency signal is omitted for easy understanding.

Reference numeral 2 denotes a pair of loudspeakers 3L arranged at the front positions substantially symmetrical with respect to the listener, as shown in FIG.
3R and a pair of speakers 3SL, 3
FIG. 3 shows signal processing means for reproducing from the SR, and the signal processing means 2 includes a pair of convolvers each having a pair of convolvers set with coefficients based on a head-related transfer function for each channel of the rear surround signals SL and SR. Right sound image localization filters 21a and 21b, similar right and left sound image localization filters 2
1c and 21d, the level of the center channel signal C is 3d
Attenuators 21e for attenuating B, adders 21f and 2 for adding the center channel signal C attenuated through the attenuators 21e to the two-channel front stereo signals L and R.
1g. FIG. 2B shows a case where a center speaker C is further arranged in FIG. 2A.

The left and right sound image localization filters 21a
Output signals of right and left and right and left sound image localization filters 2
Adder 21 for adding the output signals of 1b and 21d respectively
h and 21i, the adder 21 according to the set coefficient g _fr
Attenuators 21j and 21k for adjusting the levels of the output signals of h and 21i, and an adder for adding the rear processed signal via attenuator 21j to the output signal of adder 21f and supplying the added output to speaker 3L. 21l, similarly, adder 21g of the output signal to the attenuator 21k adds the rear processing signal through the adder supplied to the speaker 3R of the addition output 21m, the rear surround signals SL in accordance with the coefficient g _r is set attenuator 21n for supplying to the speaker 3SL to adjust the level, by adjusting the level of the rear surround signals SL in accordance with the coefficient g _r is set similarly includes an attenuator 21o to be supplied to the speaker 3SR.

Further, 4 is an input section for supplying the position information of the listener, and 5 and 6 are for adjusting and setting the level of the rear processing signal to be added to the levels of the front stereo signals L and R according to the position information. Coefficient g _fr and rear surround signal S
L, a table conversion unit and the coefficient setting unit which forms a level setting means for setting a coefficient g _r for adjusting and setting the level of SR, the table conversion unit 5, as shown solid lines in FIG. 5, for example, the listener coefficient g _fr of the attenuator 21j and 21k and 21n and 21o varies according to the position, g _r is set, the coefficient from the table conversion unit 5 based on the position information from the input unit 4 g _fr and g _r is obtained and the coefficient setting unit 6
Are set to the attenuators 21j and 21k, 21n and 21o, respectively.

Here, the rear surround signals SL, S
As shown in FIG. 3, a pair of left and right sound image localization filters 21a and 21b, 21d, and 21c provided for each channel of R include an ATAL-SCHROEDER (Atal Schrader) for canceling a spatial characteristic from a speaker. Using a type filter, processing is performed so that sound images are localized on the left and right sides of the listener, respectively, and are added to the front two-channel stereo signals L and R, respectively.

ATAL-SCHROEDER shown in FIG.
The type of filter includes two combinations of the sound image localization filter and the crosstalk canceling filter array shown in FIG.
It is configured for two sound images. That is, the configuration of FIG. 4 is obtained for one channel. The configuration of the sound image localization filter circuit shown in FIG. 4 includes a sound image localization filter for convolving localization position characteristics F and K, and a crosstalk canceling filter array, and the crosstalk canceling filter array portion is an ATAL-SCHRO.
It corresponds to an EDER filter. In FIG. 4, S represents the transfer function from a pair of speakers to the same ear, while A
Represents the transfer function to the opposite ear. Since the speakers are symmetric with respect to the listener, the transfer function from each speaker to both ears is also symmetric.

The output signals X 'from the speakers LF, RF,
Y ′ is expressed as follows: X ′ = (SX−AY) / (S ² −A ² ) Y ′ = (SY−AX) / (S ² −A ² ) To perform sound image localization, a signal obtained by convoluting a transfer function in a direction to be localized with X and Y inputs is input. That is, with input of X = Fx and Y = Kx, X ′ = [(SF−AK) / (S ² −A ² )] × x = Hl
x Y ′ = [(SK−AF) / (S ² −A ² )] × x = Hr
x, and sound image localization is realized. The filter is this Hl,
It has Hr as a coefficient.

Referring back to FIG. 3, S is a transfer function from the pair of speakers to the ear on the same side of the listener, A is a transfer function from the pair of speakers to the ear on the opposite side of the listener, and F is a sound image. Transfer function from the desired position to the listener's ear on the same side,
K indicates a transfer function from the position where the sound image is to be localized to the ear on the opposite side of the listener. Since the speakers are symmetric with respect to the listener, the transfer function from each speaker to both ears is also symmetric. The outputs from the actual speakers 3L, 3R arranged at symmetrical positions in front of the listener are X ′,
Let it be Y '. In order to localize an arbitrary sound image around the listener by the two front speakers 3L and 3R, a process of convolving transfer functions F and K from the target position to be localized to the listener with the input signal is performed, and then Speaker 3
A process for canceling the spatial characteristics from L, 3R to the listener is performed. First, the latter spatial characteristic canceling process will be described.

Outputs x, y at both ears of the speakers 3L, 3R arranged at symmetrical positions in front of the ears are represented by x = SX '+ AY' y = AX '+ SY', and the outputs are sent to the cancel filter. Input X,
Since we want to make it equal to Y, input x and y to X and Y
X = SX ′ + AY ′ Y = AX ′ + SY ′ Therefore, based on this equation, for the inputs X and Y, the outputs X ′ and Y ′ from the speaker are expressed as X ′ = ( SX−AY) / (S ² −A ² ) (1.1) Y ′ = (SY−AX) / (S ² −A ² ) (1.2)

Next, the convolution of the transfer function in the direction to be localized will be considered. In FIG. 3, when the signal x is reproduced from the speaker (position Xp) on the rear left side of the listener, the listener's left ear L
In the case of -ear and the right ear R-ear, signals Fx and Kx are obtained by convolving the transfer functions F and K, respectively. If this signal is given to the inputs X and Y, sound image localization can be realized.

That is, in FIG. 3, X = Fx, Y = K
When x is input to equation (1.2), X ′ = [(SF−AK) / (S ² −A ² )] × x = H1 × x (2.1) Y ′ = [(SK−AF) / (S ² −A ² )] · x = Hr · x (2.2) This concludes the convolution of the transfer function and the cancellation processing of the spatial characteristics into the filter coefficients Hl and Hr. This is the case where the speaker is located at the rear left position Xp of the listener, so in FIG. 3, the filter coefficient Hl indicates the filter 21a, and the filter coefficient Hr indicates the filter 21b.
And the original signal x is taken as its input. In order to form another sound image of the input signal y on the rear right side, since it is symmetrical in the same way, the filters are switched left and right and input as 21c and 21d.

Therefore, in order to localize the sound image at two positions substantially symmetrical in front of the listener, a pair of left and right sound image localization filters 21a provided for each channel of the rear surround signals SL and SR are provided. And 21b,
21d and 21c are four FIs having Hl and Hr as filter coefficients shown in the equations (2.1) and (2.2).
By using the R filter, sound image localization can be performed.

By the above operation, the decoder 1 shown in FIG.
The two-channel stereo signals L and R among the 5ch signals decoded and separated from are added to adders 21f, 21g and 21g.
Stereo speakers 3L, 3R such as televisions which are arranged at a front position substantially symmetrical with respect to the listener via the speakers 1 and 21m.
The center channel signal C is reproduced by the attenuator 21
e, after the level is dropped by 3 dB,
It is distributed to the speakers 3L and 3R via 1f and 21g.

The rear surround signals SL and SR are
A pair of left and right sound image localization filters 21a and 21b using an ATAL-SCHROEDER type filter as shown in FIG. 3 and filter processing so as to localize the sound image to the left and right sides of the listener, respectively, via 21c and 21d. After that, the adders 21h and 21i, and the attenuators 21j and 2 in which the coefficient g _fr according to the listener's position information are set.
1k, and are added to the front two-channel stereo signals L and R via adders 211 and 21m.
The listener is made to enjoy the movement sound is surrounded by the sound image of the 5ch while watching in front of the TV, while the rear surround signals SL, SR itself coefficient g _r corresponding to the position information of the listener are set Through the attenuators 21n and 21o, the sound is reproduced by the speakers 3SL and 3SR arranged at the rear position substantially symmetrical to the listener, and a high sense of realism is obtained.

That is, as shown in FIG. 2A, by providing a pair of speakers 3L and 3R on both sides of the display monitor, the level is changed from the position shown in FIG. 2B according to the position of the listener. The rear surround signals SL and SR whose are adjusted and set are reproduced, and a three-dimensional surround sound can be reproduced by a stereo signal from the front and a rear stereo signal (rear surround) localized to the rear, and a rear surround sound can be reproduced. The signal itself is reproduced by the rear speakers 3SL and 3SR.

Here, as shown in FIG. 5, speakers 3L and 3R of the front stereo signal arranged in front of
And rear speakers 3SL and 3SR arranged at rear positions
When the listener is at the center between the speakers 3L and 3R and the speakers 3SL and 3SR, the normal reproduction is performed almost as described above, but the listener is positioned at a higher position than the center. Moving forward or backward is as follows.

That is, when the listener moves forward, the listener manually inputs the position information to the input unit 4 to the input unit 4 accordingly, or obtains the distance from the four speakers using an infrared sensor or the like. When the position information is input to the input unit 4 by detecting means (not shown) for calculating the front and rear position, the table conversion unit 5 attenuates the level of the rear processed signal from zero to the enhancement direction with respect to the level of the front stereo signal. The coefficients g _fr of 21j and 21k are adjusted and set. And outside the listening area, the attenuators 21j and 21k
_Has a value of 2 and is constant. by this,
As shown by the one-dot chain line in FIG. 5, the rear sound can be clearly heard from the front speaker by adding the rear processing signal by setting the coefficient g _fr . On the other hand, when the listener moves behind the center, the level of the rear processing signal becomes zero as possible, also, to attenuate the level of the rear signal by setting the coefficient g _r attenuator 21n and 21o, speakers 3SL and The sound is adjusted so that the rear sound output from the 3SR is not noisy. As a result, the power of the rear signals output from the speakers 3SL and 3SR is attenuated, as indicated by the dashed line in FIG.

[0041] Thus, as described above, when the listener moves forward or backward from the center, attenuators 21j and 21k and 21n and the coefficient of 21o g _fr and g _r is set in accordance with the position information of moving in the longitudinal direction As a result, the sound volumes output from the speakers 3L and 3R, 3SL and 3SR are corrected, so that the sound field is corrected by a three-dimensional sound field according to the position of the listener, and the sound field is adjusted to be used in a wide range. Is done. In the above embodiment, the addition is performed by the adders 211 and 21m and supplied to the speakers 3L and 3R. However, the speakers 3L and 3R reproduce the outputs of 21f and 21j, respectively, and the two speakers 3L1 and 3L2, and 21g and 21k. May be supplied to the two speakers 3R1 and 3R2 for reproducing the output, and may be added in space, or may be added substantially. This point can be applied not only to the first embodiment but also to the following embodiments, and a pair of speakers described in the claims should be interpreted to include such a composite speaker.

<Embodiment 2> FIG. 6 is a block diagram showing an embodiment 2 of an apparatus for implementing the multi-channel signal processing method of the present invention. In the above-described first embodiment, since a pair of filters for performing a sound image localization process is required for each of the rear surround signals SL and SR for each channel, that is, four filters are required to localize the sound image to two locations. A filter is required, the hardware scale becomes large, and it cannot be used for consumer products such as a television receiver. In the second embodiment,
This is effective when sound image localization is performed at a position symmetrical with respect to the median plane where the listener's head is divided symmetrically. The number of filters to be used is half that of the first embodiment shown in FIG. By doing so, the above-mentioned problem is solved.

That is, in the second embodiment shown in FIG. 6, the signals of 5 channels decoded and separated by the decoder 1 are arranged at symmetrical positions in front of the listener, for example, for two channels of a television receiver TV. Stereo speakers 3L and 3R and rear speaker 3SL arranged at the rear position
And as signal processing means 2 for reproducing from 3SR,
The signal processing is performed so that the rear surround signals SL and SR are localized in the sound image behind the listener, and the sound image localization circuit 22a for reproducing from the front two-channel stereo speakers 3L and 3R, and the level of the center channel signal C is attenuated by 3 dB. The attenuator 22b includes adders 22c and 22d for adding the center channel signal C attenuated through the attenuator 22b to the two-channel stereo signals L and R.

An amplifier 22e for correcting the levels of the rear surround signals SL and SR via the sound image localization circuit 22a based on a coefficient g ₁ + α (= g _fr ) which is adjusted and set according to the position of the listener. And 22f, adder 22c
And 22d, the L of the rear surround signal through the sound image localization circuit 22a and the amplifiers 22e and 22f.
And R channel signal, respectively, and add
And and adder 22g and outputs the 3R 22h, listener and corrected respectively based on the coefficient 1-g ₁ be adjusted and set in accordance with the level of the rear surround signals SL and SR to the position of the listener (= g _r) Speaker 3SL located at rear position
And amplifiers 22i and 22j for outputting to the 3SR.

In the embodiment shown in FIG.
As means for inputting the coefficients g ₁ + α and 1-g ₁ to be set in the amplifiers 22e and 22f, 22i and 22j, distance detecting means 7 for detecting the distance between the listener moving back and forth and the speaker, And a position calculation / coefficient input unit 8 for inputting coefficients g ₁ + α and 1-g ₁ according to the position information based on the position of the listener.

Here, the distance detecting means 7 is not shown.
Four speakers and listeners using infrared sensors
Of the position calculation / coefficient input unit 8
Calculates the listener's front-back position based on the detected value
And a coefficient g corresponding to the position information. 1 + α and 1-g₁Amplify
Input to the devices 22e and 22f, 22i and 22j. So
At this time, as shown in FIG.
Coefficient g corresponding to the position of the listener in the front-back direction_fr(= G₁ +
α) and g_r(= 1-g₁) And associated coefficient g₁ + Α and g₁ 
Is set, and the listener moves forward from the center.
When amplifiers 22e and 22f have coefficient g₁ + Α is input
The amplifiers 22i and 22i are located behind the center.
j is the coefficient 1-g₁Is activated by inputting
You.

As the sound image localization circuit 22a,
JP-A-8-51698 which has already been proposed by the present applicant and others
(SHUFFLER)
Use a filter. The configuration of this shuffler filter is shown in FIG.
Shown in In FIG. 8, 22a ₁ is a first adder for obtaining a sum signal of the left and right rear surround signals, 22a ₂
The processes of the second adder for obtaining a difference signal of the left and right pair of rear surround signals, 22a ₃ the output of the adder 22a ₁ filter coefficient P is set first to be described later 1
Filters, 22a ₄ second for processing an output of the second adder 22a ₂ are set filter coefficient N to be described later
22a ₅ is a third adder that obtains a sum signal of the signals processed by the first and second filters, and 22a ₆ is a signal that calculates the difference signal between the signals processed by the first and second filters. A fourth adder to obtain the third and fourth adders 22.
the output of a ₅ and 22a ₆ and to output a filtered pair of left and right rear surround signals.

Here, when the equations (1.1) and (1.2) are modified, X '= [(XY) / (SA)] + Y' Y '= [(X + Y) / (S + A )] − X ′ and solving for the output signals X ′ and Y ′, X ′ = N (XY) + P (X + Y) Y ′ = P (X + Y) -N (XY) It is shown that a hula filter can be configured,
The filter coefficients P and N are given by P = 1 / (S + A) (3.1) N = 1 / (S−A) (3.2) The output signal X ', Y' is, X '= [2 (SX-AY)] ^{/ (S 2 -A 2) (} 4.1) Y' = [2 (SY-AX)] / (S ² −A ² ) (4.2), and a result similar to the expressions (1.1) and (1.2) of the first embodiment is obtained. However, since it has a double value, it has a gain of 6 dB.

Then, the numerators of the equations (3.1) and (3.2) are expressed as follows: P = (F + K) / (S + A) (5.1) S = (F−K) / (SA) (5. if 2), X '= [2 (SFX + SKY-AFY- AKX) ] ^{/ (S 2 -A 2) (} 6.1) Y' = [2 (SFY + SKX-AFX- AKY) ] / (S ² - A ² ) (6.2) is obtained.

When the input X = x and Y = 0, the same result as in the equations (2.1) and (2.2) can be obtained.
On the other hand, for an input of X = 0 and Y = y, X ′ = Hr · y (7.1) Y ′ = Hl · y (7.2) This indicates that when input to y, the sound image is localized at a position symmetrical to the sound image position input to x, and when signals are input to x and y, the superposition principle holds, The sound images are localized at symmetrical positions.

Therefore, according to the second embodiment,
The sound image localization circuit 22a for the rear surround signals SL and SR
And use a shuffler filter to create sound on the left and right sides of the listener.
Perform processing to localize the image, and adjust the position according to the position of the listener.
Several g_frAre adjusted and set via amplifiers 22e and 22f.
And add it to the front two-channel stereo signals L and R
And the rear surround signals SL and SR are
Coefficient g according to the position of amplifier 22 where r is adjusted and set
By outputting via i and 22j,
As shown in FIG. 2 (A), the listener listens to the television
Move while being surrounded by 5ch sound image while watching TV
You can enjoy the sound and get a high sense of reality
The number of filters to be used is different from that of the first embodiment shown in FIG.
Only two halves are required, and the configuration can be simplified.
Reduce the cost and incorporate it into consumer television receivers.
Both will be possible. Also, at that time, depending on the position of the listener
The sound field is corrected over a wide range by performing sound field correction using a three-dimensional sound field.
Can be adjusted for use in the box.

In the description of each of the above embodiments, the AC-3 system developed by Dolby Laboratories is used as an example of a 5ch audio signal. However, the present invention is not limited to this. SDDS (Sony Dynamic Digital) divided into 2 channels of L and R
Sound), DTS (Digital Theatre)
er Sound).

<Embodiment 3> Next, FIG.
Embodiment of an apparatus for implementing a multi-channel signal processing method
FIG. 9 is a configuration diagram illustrating a state 3. With the second embodiment shown in FIG.
If only the differences are described, in Embodiment 3 shown in FIG.
Are between the adders 22c and 22g and between the adders 22d and 2d.
2h and amplifiers 22k and 221 respectively
And the coefficient g for the amplifiers 22e and 22f 1 (=
g_fr) According to the setting of the amplifiers 22k and 22l.
Number 1-g₁(= G_f) Is set. Sand
That is, the speaker 3L and the front of the position detection / coefficient input unit 9
And 3R related amplifiers 22e and 22f and amplifier 22k and
And the coefficient g of 22l₁And 1-g₁Automatically defines the coefficient g
_fAnd g_frAdjust the volume while keeping the volume
I have. The coefficient related to the rear speakers 3SL and 3SR is one.
Fixed. Back speakers 3SL and 3SR and listener
Is far enough and high, with
If the sound behind does not feel loud,
The level of the rear speaker may be constant.

In the first to third embodiments, the level of the pair of left and right rear processed signals to be added to the front stereo signal is adjusted and set according to the position of the listener, and the level of the rear processed signal is adjusted and set. In the fourth to sixth embodiments described below, the level of a pair of left and right front processed signals to be added to the rear signal is adjusted and set according to the position of the listener. The invention for adjusting and setting the level of the forward signal according to the adjustment and setting level of the forward processing signal will be described.

In other words, the multi-channel signal processing method and apparatus according to another invention comprises a pair of front left and right speakers disposed at a front position and a pair of rear speakers disposed substantially symmetrically to the listener. A multi-channel signal processing method for reproducing a multi-channel signal including a stereo signal and a rear signal, the filter realizing a convolver in which a filter coefficient based on a head-related transfer function is set for the front stereo signal A filtering step and a means for outputting a pair of processed left and right front processed signals; an adding step and a means for adding the pair of left and right front processed signals filtered by the filtering step to the left and right rear signals; A second step of adjusting and setting the level of the pair of left and right front processing signals to be added to the rear signal And level setting step and means,
A second level setting step and means for adjusting and setting the level of the front stereo signal according to the adjustment and setting level of the front processing signal.

<Embodiment 4> FIG. 10 is a block diagram showing a fourth embodiment of an apparatus for implementing a multi-channel signal processing method according to another invention. In the configuration shown in FIG. 10, the same portions as those in Embodiment 1 shown in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted. Adders 21f and 21g
Each subtractor 21e front stereo signals L and R of the center channel signal C are added attenuated through the amplifier to adjust and set the output level in accordance with the coefficient g _f which is set according to the position of the listener 21p by And 21q
2A, the signals are supplied to a pair of speakers 3L and 3R arranged at a front position substantially symmetrical to the listener as shown in FIG.

The front stereo signals L and R via the adders 21f and 21g are input to the left and right sound image localization filters 21a and 21b and 21c and 21d, respectively, and are added to the left and right by the adders 21h and 21i. The output signals of the sound image localization filters 21a and 21c are added, and the right and left sound image localization filters 21b and 21c are added.
The output signals of d are added, and the output levels of the added outputs are adjusted by attenuators 21j and 21k according to a coefficient _grf set according to the position of the listener. Rear surround signals SL and SR output from decoder 1
Is transmitted to the attenuator 21j via adders 21r and 21s.
And a pair of left and right front processing signals output from the speaker 3S and the speaker 3S as shown in FIG.
L and 3SR.

Here, a pair of left and right sound image localization filters 21a and 21b, 21d and 21c provided for each channel of the front stereo signals L and R, as shown in FIG. ATAL-SCHROEDER (Atal
Using a (Schrader) type filter, processing is performed so that sound images are localized on the left and right sides of the listener, respectively, and added to the rear surround signals SL and SR.

ATAL-SCHROEDE shown in FIG.
The R-type filter has a combination of the sound image localization filter and the crosstalk canceling filter array shown in FIG. 12 configured for two sound images. That is, the structure of FIG. 12 is obtained for one channel. The configuration of the sound image localization filter circuit of FIG.
And a crosstalk canceling filter array, and the crosstalk canceling filter array portion is an ATAL-SC
It corresponds to an HROEDER filter. In FIG.
S represents the transfer function from a pair of speakers to the same ear, while A represents the transfer function to the opposite ear. Since the speakers are symmetric with respect to the listener, the transfer function from each speaker to both ears is also symmetric.

[0060] Speaker 3SL, the output signal X from the 3SR ', Y' is, X '= (SX-AY ) / (S 2 -A 2) Y' = (SY-AX) / (S 2 -A 2) Becomes To perform sound image localization, a signal obtained by convoluting a transfer function in a direction to be localized with X and Y inputs is input. That is, with input of X = Fx and Y = Kx, X ′ = [(SF−AK) / (S ² −A ² )] × x = Hl
x Y ′ = [(SK−AF) / (S ² −A ² )] × x = Hr
x, and sound image localization is realized. The filter is this Hl,
It has Hr as a coefficient.

Returning to FIG. 11, S is a transfer function from the pair of speakers to the ear on the same side of the listener, A is a transfer function from the pair of speakers to the ear on the opposite side of the listener, and F is a sound image. K indicates a transfer function from a desired position to the ear on the same side of the listener, and K indicates a transfer function from a position where the sound image is to be localized to the ear on the opposite side of the listener. Since the speakers are symmetric with respect to the listener, the transfer function from each speaker to both ears is also symmetric. The outputs from the actual speakers 3SL and 3SR arranged at symmetrical positions behind the listener are X 'and Y'. Rear two speakers 3SL, 3
In order to localize an arbitrary sound image around the listener by the SR, a process of convolving transfer functions F and K from the target position to be localized to the listener to the input signal is performed, and then receiving from the rear speakers 3SL and 3SR. A process for canceling the spatial characteristics up to the listener may be performed. As in the first embodiment, the output of the speakers 3SL and 3SR is expressed by the formula (2.
1) and (2.2) can be obtained, and convolution of the transfer function and cancellation processing of the spatial characteristics are combined into the filter coefficients Hl and Hr.

Accordingly, in order to localize the sound image at two positions substantially symmetrical behind the listener, a pair of left and right sound image localization filters 21a provided for each channel of the front stereo signals L and R are provided. And 21b, 21d
And 21c are composed of four FIR filters having Hl and Hr as filter coefficients shown in the equations (2.1) and (2.2), thereby enabling sound image localization.

[0063] The above operation, among 5ch signals decoded separated from decoder 1 shown in FIG. 10, two-channel stereo signal L, R is an adder 21f, 21g, the coefficient g _f corresponding to the position of the listener Amplifier 21 to be set
Stereo speakers 3L, 3R such as a television arranged at a front position substantially symmetrical to the listener via p and 21q.
The center channel signal C is reproduced by the attenuator 21
e, after the level is dropped by 3 dB,
f, 21g, the speakers 3L and 3R and the speaker 3S arranged at a rear position substantially symmetrical to the listener.
L and 3SR. Further, the rear surround signals SL and SR are attenuators 21j and 21k in which adders 21r and 21s set a coefficient _grf according to the position information.
Are added to the front two-channel stereo signals L and R subjected to the sound image localization processing via the
Thus, the listener can enjoy the moving sound while being surrounded by the sound image of 5ch while watching the television in front.

Here, the coefficient g _rf of the attenuators 21j and 21k and the coefficient g _f of the amplifiers 21p and 21q change as the listener moves forward or backward from the center as shown by the solid line in FIG. Is set. Attenuator 21
coefficients g for j and 21k and amplifiers 21p and 21q
_rf and g _f applied before the front sound output from the rear speakers 3SL and 3SR disposed on the speakers 3L and 3R and the rear position of the front stereo signal disposed in front and rear sound intensity level (power) 13 When the listener is at the center between the speakers 3L and 3R and the speakers 3SL and 3SR, almost normal reproduction is performed. That is, the coefficient g _rf set for the attenuators 21j and 21k is substantially zero, and the coefficient g _f set for the amplifiers 21p and 21q is 1.

When the listener moves backward from the center, the position information is manually input to the input unit 4 accordingly, or the distance to the four speakers is calculated using an infrared sensor or the like to calculate the front / rear position (not shown). By inputting the position information to the input unit 4 by the detecting means, the table conversion unit 5 adjusts the level of the front processed signal from zero to the level of the rear sound (rear surround signal) from zero to an attenuator. set the coefficient g _rf of 21j and 21k, the coefficient g _rf attenuator 21j and 21k is a value of 2 becomes constant in the non-listening area. Then, the level of the rear surround signal is attenuated in accordance with the increase in the level of the forward processing signal,
Adjust so that one is balanced.

As a result, a dashed line is shown in FIG.
Like the coefficient g from the rear speaker r_fForward processing by setting
The forward sound can be clearly heard with the addition of the logical signal. one
On the other hand, if the listener moves forward from the center,
The level of the signal becomes zero without limit, and the amplifier 21
p and coefficient g of 21q_f The level of the forward signal
Before output from speakers 3L and 3R
Adjusted so that the tone is not too loud. By this
As shown by the dashed line in FIG.
The power of the forward signal output from the 3R is attenuated.

Accordingly, as described above, when the listener moves forward or backward from the center, the attenuators 21j and 21k and the amplifier 21
The coefficients g _rf and g _{f of} p and 21q are set, and the volumes output from the speakers 3L and 3R, 3SL and 3SR are corrected. As a result, the sound field is corrected by the three-dimensional sound field according to the position of the listener. The sound field is adjusted so that it can be used in a wide range.

<Fifth Embodiment> FIG. 14 is a block diagram showing a fifth embodiment of an apparatus for performing a multi-channel signal processing method according to another invention. In the above-described fourth embodiment, a filter for performing a pair of sound image localization processes for each of the front stereo signals L and R is required for each channel. In other words, four filters are required to localize the sound image to two locations. A filter is required, the hardware scale becomes large, and it cannot be used for consumer products such as a television receiver. This fifth embodiment is effective when sound image localization is performed at a position symmetrical with respect to the median plane that divides the listener's head symmetrically, and the number of filters to be used is determined according to the fourth embodiment shown in FIG. Half of 2
The configuration described above solves the above-mentioned problem.

That is, in the embodiment shown in FIG. 14, the 5-channel signal decoded and separated by the decoder 1 is placed in a symmetrical position in front of the listener, for example, a 2-channel stereo speaker of a television receiver TV. 3
L, 3R and signal processing means 2 for reproducing from the rear speakers 3SL and 3SR arranged at the rear position, so that the front stereo signals L and R are signal-processed so as to localize the sound image to the rear of the listener, and the rear two channels are processed. Sound image localization circuit 2 for reproducing from stereo speakers 3SL and 3SR
2a, an attenuator 22b for attenuating the level of the center channel signal C by 3 dB, and a center channel signal C attenuated via the attenuator 22b into a two-channel stereo signal L.
It has adders 22c and 22d for adding to R.

An amplifier 22e for correcting the levels of the front stereo signals L and R via the sound image localization circuit 22a based on a coefficient g ₁ + α (= g _rf ) which is adjusted and set according to the position of the listener. And 22f, adders 22c and 2
A speaker 3 in which the signal via 2d is amplified based on a coefficient 1-g ₁ (= g _f ) which is adjusted and set according to the position of the listener, and the amplified output is arranged at a position in front of the listener.
Amplifiers 22k and 22l for feeding L and 3R;
Speakers 3SL and 3SR that are added to the rear surround signals SL and SR to the L and R channel signals of the front stereo signal via the amplifier 22e and arranged at the rear position of the listener
Are provided with adders 22m and 22n.

In the fifth embodiment shown in FIG. 14, the means for inputting the coefficients g ₁ + α and 1-g ₁ to be set in the amplifiers 22e and 22f, 22k and 22l is used as a means for moving the listener back and forth. A distance detecting means 7 for detecting the distance between the speaker and the speaker; a position calculating / coefficient inputting unit 8 for calculating the position of the listener based on the detected value and inputting the coefficients g ₁ + α and 1-g ₁ according to the position information; Have.

Here, the distance detecting means 7 is not shown.
Four speakers and listeners using infrared sensors
Of the position calculation / coefficient input unit 8
Calculates the listener's front-back position based on the detected value
And a coefficient g corresponding to the position information. 1 + α and 1-g₁Amplify
Input to the containers 22e and 22f, 22k and 22l. So
At this time, as shown in FIG.
Coefficient g corresponding to the position of the listener in the front-back direction_rf(= G
₁ + Α) and g_f(= 1-g₁) And associated coefficient g₁ + Α and
g₁ Is set, and the listener moves
The amplifiers 22e and 22f have the coefficient g₁ + Α is input
And when it is ahead of the center, the amplifier 22k and
Coefficient 1-g for 22l₁Is activated by inputting
I have.

As the sound image localization circuit 22a,
JP-A-8-51698 which has already been proposed by the present applicant and others
(SHUFFLER)
Use a filter. As shown in FIG. 16, the configuration of this shuffler filter has the same configuration as that of the second embodiment shown in FIG. 8, and has filter coefficients P and N shown in equations (5.1) and (5.2). , and then outputs from the third and the speaker 3SL and 3SR and the output of the fourth adder 22a ₅ and 22a ₆ placed listener rear position as filtered pair of left and right front stereo signal.

Therefore, according to the fifth embodiment,
Using a shuffler filter as the sound image localization circuit 22a for the front stereo signals L and R, processing is performed so that the sound image is localized on the left and right sides of the listener, and a coefficient g _rf corresponding to the position of the listener
After the amplifiers 22e and 22f are adjusted and set, the rear two-channel rear surround signals SL and SR
And outputs the front stereo signals L and R.
By to output via amplifier 22k and 22l of the coefficient g _f is adjusted and set in accordance with the position of the listener, the listener, the sound image of 5ch while watching in front of the television receiver TV In addition to being able to enjoy the sound of the movement being enclosed, the number of filters to be used can be reduced to half that of the fourth embodiment shown in FIG. 10, and the configuration can be simplified and the cost can be reduced. It can also be incorporated into consumer television receivers. Also,
At this time, the sound field can be adjusted so that the sound field can be used in a wide range by performing the sound field correction based on the three-dimensional sound field according to the position of the listener.

In the above embodiments, the AC-3 system developed by Dolby Laboratories is used as an example of the 5ch audio signal. However, the present invention is not limited to this, and the surround signal is not limited to this. It includes all signal formats divided into L and R channels.

<Sixth Embodiment> FIG. 17 is a block diagram showing a sixth embodiment of an apparatus for performing a multi-channel signal processing method according to another invention. Describing only difference from the fifth embodiment shown in FIG. 14, in the sixth embodiment shown in FIG. 17, a rear surround via amplifiers 22i and 22j coefficient g _r corresponding to the position of the listener are set The signals SL and SR are filtered through amplifiers 22e and 22f to form a forward stereo signal and adders 22m and 2m.
2S, the speaker 3S arranged at the rear position
L and 3SR. That is, the amplifiers 22e and 22f related to the front stereo signals L and R and the rear surround signal SL from the position detection / coefficient input unit 9.
And the coefficients g ₁ and 1 of the amplifiers 22i and 22j related to SR
Defines a -g ₁ are automatically coefficients g _rf and g _r is changed while the sound volume to surrounding constant. The coefficients relating to the front speakers 3L and 3R are assumed to be constant. In the case where the distance between the front speakers 3L and 3R and the listener is sufficiently large and high, and the front sound does not feel noisy even when moving forward, the level of the front speakers is fixed in this way. It may be.

<Embodiment 7> By the way, recently, personal computers (hereinafter, referred to as PCs or personal computers) have rapidly become multimedia, and thus moving images, sounds, and the like have been widely used by PCs. . In the seventh embodiment, a case where the above-described multi-channel signal processing method is implemented by a personal computer will be described with reference to a configuration diagram shown in FIG. As shown in FIG. 18, a personal computer 16 to which a surround signal processing program and a surround source are supplied via a disk drive 14 or a network (for example, the Internet) terminal 15
Is a CPU (equivalent to a controller) equipped with MMX (extended instruction set of P55C by INTEL: an instruction set for specific applications and added mainly to efficiently handle digital signal processing such as image and audio).
16a and RA used as a buffer during data processing.
M16b, a data converter 16c for converting data input via the disk drive 14 or the network terminal 15, and an audio interface for supplying a signal processing result to a plurality of speakers 3L, 3R, 3SL and 3SR via an amplifier. 16d.

Next, the operation of the configuration shown in FIG. 18 will be described with reference to the flowchart shown in FIG. FIG.
Indicates a state in which a disk on which a surround signal processing program is recorded is set in the disk drive 14, and a program load command (command) is input from a keyboard (not shown).
At step S1, it is determined that the program is loaded (when a command is input, it is determined that the program is loaded). CPU 16a reads out the program data
(Step S2), and when the program loading is completed, the program loading flag is set (step S3) and the process is terminated. At this time, the CPU 16
Since a is compatible with MMX, signal processing can be performed at high speed.

Next, a play command is input from a keyboard (not shown) in a state in which a disk on which a surround source is recorded is set in the disk drive 14, and when starting in this state, the program is read in step S1 by checking the program load flag. Since it is known that the disc is set (NO in step S1), the process proceeds to step S4, in which the first track (special track) of the disc is accessed to read a subcode representing the type of the disc. J
If it is ES, the data is read from the next track in step S5, decoded, compressed / limited, processed data is set in the audio interface 16d (step S6), and the process returns to step S5. Press / limit processing is repeated. The audio interface 16d performs D / A conversion on the received data at a predetermined sampling period, converts the data into analog signals, and supplies the analog signals to the speakers 3L, 3R, 3SL, and 3SR. If "NO" in the step S4, the display indicates that the performance cannot be performed (step S7), and ends.

Here, the signal processing program executed in the decoding processing in step S5 described above is composed of the processing steps shown in FIG. That is, step S31
Performs decoding processing of a multi-channel audio signal including a pair of left and right rear surround signals in the AC-3 system, and in a subsequent step S32, a filter coefficient based on a head-related transfer function is calculated for each channel of the pair of left and right rear surround signals. Filter processing for localizing the sound image to the listener in a substantially symmetrical rear position is performed by the filter processing for realizing the set convolver. In step S33, a center addition process is performed in which the front center channel signal of the multi-channel audio signal is attenuated by a predetermined amount and added to the pair of left and right front stereo signals, and in step S34, the position calculation shown in FIGS. / Coefficient input unit 8 (position detection / coefficient input unit 9 in FIGS. 9 and 17) to obtain the coefficient g ₁ +
Read α, 1−g ₁ and set it to the attenuator. afterwards,
A left / right addition process of adding a pair of left and right rear surround signals via the filtering process step S32 and the subtraction process step S34 to the pair of left and right front stereo signals to which the center channel signal has been added in step S35 is performed. An output obtained by localizing the sound image at a rear position substantially symmetric to the listener obtained through step S35 is output to the audio interface 16d shown in FIG. 18 via the output step S36.

In the AC-3 encoder, as shown in FIG. 21, the frequency of the input signal is analyzed by the filter bank 80 (for analysis), and the signal quantized by the quantizer 81 is sent to the multiplexer 82. On the other hand, the input signal is also supplied to the bit allocator 83. The output of the filter bank 80 is also provided to a spectrum envelope encoder 84.
The output of the spectrum envelope encoder 84 (which differentially encodes the spectrum envelope in the frequency domain) is output to a bit allocator 83, a core bit allocator 85, and a multiplexer 82.
Supplied to The output (bit assignment sub-information) of the bit allocator 83 is supplied to the bit allocator 85 and the multiplexer 82 of the core. The output of the core bit allocator is supplied to a quantizer 81. The multiplexer 82 outputs an encoded bit stream. With such a configuration, psychoacoustics taking account of auditory masking characteristics is adopted for bit allocation from the spectrum envelope to the core, and the bit allocation unit 83 calculates the theoretically most accurate bit allocation information and sends it to the multiplexer 82. Therefore, adaptive bit allocation with high compression efficiency can be performed by the multiplexer 82.

Such a bit stream is composed of 31.
It consists of a frame of 25 Hz (32 ms). Each frame has synchronization information (sy) composed of a 16-bit synchronization word and 8 bits indicating a sampling frequency and a frame size.
nc info), BSI (bit stream information: number of coded channels, etc.), 6 conversion blocks (signal information), and unused data (auxiliary data)
Consists of

As shown in FIG. 22, the step S31 for decoding the signal encoded in this way is, specifically, a demultiplexer 90, a spectrum envelope decoder 91, and an inverse quantizer. , A core bit allocator 93, and a filter bank (for synthesis) 94. The resolution of the filter bank 94 is variable.

In the demultiplexer 90, the state of the decoder is set from the information of the frame and the signal information of the frame is decoded by the procedure shown in FIG. The original signal (multi-channel) is reproduced by a procedure reverse to the encoding.

By recording a surround signal processing program having such processing steps on a recording medium such as a disk, it is possible to easily perform surround source signal processing according to the surround signal processing program from the recording medium. Thus, AC-3 compatible computer software can be provided. As described above, when the surround signal processing program and the surround source are supplied as software, the MMX compatible CP
U16a enables high-speed signal processing.

Embodiment 8 Further, as a recording medium such as a disk, a plurality of tracks are formed on a recording surface,
And a disk in which each track is divided into a plurality of sectors and each sector is assigned a unique address, and on a recording surface thereof, a first data area in which the surround signal processing program is recorded in a plurality of sectors; A second data area in which a surround source is recorded in a sector different from the plurality of sectors, and a TOC (TABLE) in which data indicating that the first and second data areas are present on a recording surface are recorded.
OF CONTENTS) area, the surround signal processing program is recorded as the first data, and the surround source is recorded as the second data, so that both the surround signal processing program and the surround source are recorded. AC-3 compliant computer software can be provided.

FIG. 23 is a diagram schematically showing a plane of a data-recorded disk shown as a preferred embodiment of the recording medium. The disk 17 shown in FIG. 23 is, for example, an optical recording as a DVD audio disk on which an audio signal having a diameter of 120 mm specified by a standard disk also referred to as CD Plus (CD PLUS in System Identifiier) or enhanced music CD is recorded. A first lead-in area 17a, a first data area 17b, a first lead-out area 17c, a second lead-in area 17d, a second data area 17e, and a second lead-out area 17f from the inner circumference toward the outer circumference;
Are arranged substantially concentrically in this order. here,
The first lead-in area 17a is the first data area 17
First TOC for recording the address of each data of b
And the second lead-in area 17d is
A second TOC for recording an address of each data in the data area 17e is formed.

The operation of the CPU 16a shown in FIG. 18 when the disk shown in FIG. 23 is used as a recording medium will be described with reference to the flowchart shown in FIG. FIG.
Composite disk 17 having two data areas as shown in FIG.
Is set in the disk drive 14, when a play instruction is input from a keyboard (not shown), it is first determined whether or not the program has been loaded by checking a flag (step S11). 23, the program data area of the CPU is known by looking at the first lead-in area 17a of the inner first data area 17b shown in FIG.
The programs recorded in the first data area 17b are sequentially loaded into the internal RAM of the CPU, and continue until the program loading ends (steps S12 to S15). If it is determined that the program is absent (step S
13 (YES), ends.

When the program loading is completed (YES in step S15), the test data stored after the first data program is read out and loaded into the RAM 16b (step S16). Not performed). Then, a test decoding process is performed by operating the decoding program (S17). As a result, when it is determined that the data has been correctly decoded (because a correct decoded signal is supplied as a pair with the test data in advance, it is determined whether or not the data is correct by comparing them), and the program loading is completed. Is set (S19), and the process returns to step S11. If "NO" in the step S18, the performance disable is displayed on the display and the process ends. Note that AC-
The present invention can be applied not only to 3 but also to surrounds of other standards. If YES is determined in step S11, that is, if the program has been loaded, the second data recorded in the second data area 17e is read (read), and the second data is read into the end of file (EOF). The same decoding process as described above is performed until it is output to the audio interface 16d (steps S21 to S2).
4). If YES in step S22, the process ends.

[0090]

As described above, according to the multi-channel signal processing method and apparatus according to the present invention, the sound supplied to the front or rear of the sound reproduced by the front speaker and the rear speaker is determined by the position of the listener. 3D sound field processing is applied to the rear speaker or the front speaker according to the sound field, so that the optimum sound field range is expanded, so that the sound field correction by the 3D sound field is performed according to the position of the listener. There is an effect that it can be adjusted so that it can be used in a wide range.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing Embodiment 1 of an apparatus for implementing a multi-channel signal processing method of the present invention.

FIG. 2 is an explanatory diagram showing an arrangement of a speaker according to the present invention.

3 is a filter (ATAL-SCHR) used as a sound image localization circuit for the rear surround signals SL and SR in FIG.
FIG. 3 is a configuration diagram of an OEDER-type filter.

FIG. 4 is a configuration diagram of a sound image localization circuit including an ATAL-SCHROEDER filter as a component of the ATAL-SCHROEDER type filter of FIG. 3;

FIG. 5 is an explanatory diagram of coefficients and output power of the attenuator of FIG. 1;

FIG. 6 is a configuration diagram showing Embodiment 2 of an apparatus for performing a multi-channel signal processing method of the present invention.

FIG. 7 is an explanatory diagram relating to coefficients of the amplifier in FIG. 6;

8 is a filter (shuffler filter) used as a sound image localization circuit for the rear surround signals SL and SR in FIG.
FIG.

FIG. 9 is a configuration diagram showing Embodiment 3 of an apparatus for performing a multi-channel signal processing method of the present invention.

FIG. 10 is a configuration diagram showing Embodiment 4 of an apparatus for performing a multi-channel signal processing method of the present invention.

11 is a filter (ATAL-SCHROEDER) used as a sound image localization circuit for the front signals L and R in FIG.
FIG.

12 is an ATAL-SCHROEDE as a component of the ATAL-SCHROEDER type filter of FIG. 11;
FIG. 3 is a configuration diagram of a sound image localization circuit including an R filter.

FIG. 13 is an explanatory diagram of coefficients and output powers of the attenuator and the amplifier of FIG. 1;

FIG. 14 is a configuration diagram showing Embodiment 5 of an apparatus for performing the multi-channel signal processing method of the present invention.

FIG. 15 is an explanatory diagram relating to coefficients of the amplifier in FIG. 14;

16 is a configuration diagram of a filter (shuffler filter) used as a sound image localization circuit for the front signals L and R in FIG.

FIG. 17 is a configuration diagram showing Embodiment 6 of an apparatus for performing the multi-channel signal processing method of the present invention.

FIG. 18 is a configuration diagram according to an embodiment in a case where a multi-channel signal processing method according to the present invention is implemented by a personal computer as an apparatus.

19 is a flowchart showing an operation in the configuration shown in FIG.

FIG. 20 is a flowchart showing processing steps of a multi-channel signal processing program recorded on a recording medium.

FIG. 21 is a block diagram illustrating a configuration of an AC-3 encoder.

FIG. 22 is a block diagram of a configuration for realizing a step of decoding an encoded signal.

FIG. 23 is a diagram schematically showing a plane of a data recorded disk used as a recording medium.

24 is a flowchart showing the operation of the CPU 16a shown in FIG. 18 when the disk shown in FIG. 23 is used as a recording medium.

FIG. 25 is an explanatory diagram relating to a relationship between a listener position and output power according to a conventional example.

[Explanation of symbols]

Reference Signs List 1 decoder 2 signal processing unit 3L, 3R, 3SL, 3SR speaker 4 input unit 5 table conversion unit 6 coefficient setting unit 7 distance detection unit 8, 9 position calculation / coefficient input unit 14 disk drive 15 network terminal 16a CPU 16b RAM 16d audio Interface 17 Disk (recording medium) 17a First lead-in area (TOC area) 17b First data area 17d Second lead-in area (TOC area) 17e Second data area 21a, 21d Left sound image localization filter (sound image localization means) 21b , 21c Right sound image localization filter (sound image localization means) 21e, 21j, 21k, 21n, 21o, 22b
Attenuator 21f, 21g, 21h, 21i, 211, 21m, 2
1r, 21s, 22c, 22d, 22g, 22h, 22
m, 22n Adders 21p, 21q, 22e, 22f, 22i, 22j, 2
2k, 22l Amplifier 22a Shuffler filter (sound image localization circuit) 22a1 First adder 22a2 Second adder 22a3 First filter 22a4 Second filter 22a5 Third adder 22a6 Adder for arithmetic 4 S31 AC- 3 decoding processing step S32 Filter processing step S33 Center addition processing step S34 Left and right addition processing steps 80, 94 Filter bank 81 Quantizer 82 Multiplexer 83 Bit allocator 84 Spectrum envelope encoder 85, 93 Core bit allocator Reference Signs List 90 Demultiplexer 91 Decoder of spectrum envelope 92 Dequantizer

Claims (14)

[Claims] 1. A multi-channel signal including a pair of left and right front stereo signals and a rear signal is reproduced from a pair of speakers disposed at a front position and a pair of speakers disposed at a rear position substantially symmetrically to a listener. A multi-channel signal processing method, comprising: outputting a pair of filtered left and right rear processed signals for realizing a convolver in which a filter coefficient based on a head-related transfer function is set for the rear signal. An adding step of substantially adding the pair of left and right rear processed signals filtered by the filtering step to the left and right front stereo signals; and a level of the pair of left and right rear processed signals to be added to the front stereo signals. A first level setting step of adjusting and setting the level, and an adjusting and setting level of the rear processing signal Accordance with the second level setting step of adjusting and setting the level of the rear signal, multi-channel signal processing method comprising. 2. A step of supplying an output signal of the adding step to a pair of speakers disposed at the front position, and a step of supplying a rear signal having passed through the second level setting step to a pair of speakers disposed at the rear position. 2. The multi-channel signal processing method according to claim 1, further comprising the step of: 3. The method according to claim 2, further comprising the step of: providing position information for supplying position information of a listener, wherein said first level setting step includes the step of adding the level of the stereo processing signal to the level of the stereo signal in accordance with the position information. 3. The multi-channel signal processing method according to claim 1, wherein the level is adjusted and set. 4. The multi-channel signal is a signal of a predetermined standard format of a 5-channel configuration of a 2-channel front stereo signal, a 1-channel center channel signal, and a 2-channel rear signal. 4. The multi-function device according to claim 1, further comprising a center addition step of attenuating the center channel signal and adding the attenuated center channel signal to a pair of left and right front stereo signals. Channel signal processing method. 5. The multi-channel signal is a wideband audio signal recorded on a DVD audio disk, and is a decoding device that decodes a signal recorded on the DVD audio disk before the filtering step and the adding step. 5. The multi-channel signal processing method according to claim 1, further comprising a processing step. 6. The filter processing step comprises the steps of: filtering the filter coefficients Hl and H for each channel of a pair of left and right rear signals;
r is Hl = (SF-AK) / (S 2 -A 2) Hr = (SK-AF) / (S 2 -A 2) ( however, S is from a pair of speakers to the same side of the ear of the listener A is a transfer function from a pair of speakers to the ear on the opposite side of the listener, F is a transfer function from a position where the sound image is to be localized to the ear on the same side of the listener, and K is a position where the sound image is to be localized. Transfer function from the ear to the other ear of the listener)
, And adds the filtered output with the filter coefficient set to Hl and the filtered output with the different filter coefficient Hr of the other channel based on the set filtered output, respectively, and filters the pair of added outputs. 6. The multi-channel signal processing method according to claim 1, wherein the signal is output as a pair of left and right rear processed signals. 7. The filter processing step, wherein a filter coefficient P is P = (F + K) / (S + A) (where S is a transfer function from a pair of speakers to an ear on the same side of the listener, and A is a pair of speakers , The transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the position where the sound image is to be localized to the opposite ear of the listener. Transfer function up to)
And the first filter processing output obtained by filtering the sum signal of the pair of left and right rear signals, and the filter coefficient N is set to N = (F−K) / (S−A). A sum signal and a difference signal of the first and second filtered outputs based on a second filtered output obtained by filtering the difference signal of 7. The multi-channel signal processing method according to claim 1, wherein the signal is output as a rear-processed signal. 8. A multi-channel signal including a pair of left and right front stereo signals and a rear signal is reproduced from a pair of speakers disposed at a front position substantially symmetrically with respect to a listener and a pair of speakers disposed at a rear position. A multi-channel signal processing method, comprising: outputting a pair of filtered left and right forward processed signals for realizing a convolver in which a filter coefficient based on a head-related transfer function is set for the front stereo signal. And an adding step of adding the pair of left and right front processed signals filtered by the filtering step to the left and right rear signals; and adjusting and setting the level of the left and right pair of front processed signals to be added to the rear signal. A first level setting step of adjusting the forward processing signal according to the adjustment / setting level of the forward processing signal. A second level setting step of adjusting and setting the level of the stereo signal, a multi-channel signal processing method comprising. 9. A step of supplying an output signal of the adding step to a pair of speakers arranged at the rear position, and a step of supplying a front stereo signal having passed through the second level setting step to the pair of speakers arranged at the front position. The method of claim 8, further comprising the step of providing. 10. A position information supply step of supplying position information of a listener, wherein the first level setting step includes the step of adding the front processing signal to the level of the rear signal in accordance with the position information. 10. The multi-channel signal processing method according to claim 8, wherein the level is adjusted and set. 11. The multi-channel signal is a 5-channel AC-3 type signal including a front stereo signal of 2 channels, a center channel signal of 1 channel, and a rear signal of 2 channels, and the second level setting. 11. The method according to claim 8, further comprising, before the step, a center adding step of attenuating the center channel signal to substantially add the center channel signal to a pair of left and right front stereo signals. Multi-channel signal processing method. 12. The multi-channel signal is a DVD
9. The method according to claim 8, further comprising a decoding step of decoding a signal recorded on the DVD audio disk before the filtering step and the adding step, which is a wideband audio signal recorded on an audio disk. 12. The multi-channel signal processing method according to any one of 8 to 11. 13. The filter processing step is such that, for each channel of a pair of left and right front stereo signals, the filter coefficients Hl and Hr are Hl = (SF−AK) / (S ² −A ² ) Hr = (SK−AF) ) / (S ² −A ² ) (where S is the transfer function from the pair of speakers to the ear on the same side of the listener, A is the transfer function from the pair of speakers to the ear on the opposite side of the listener, and F is Transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, K is the transfer function from the position where the sound image is to be localized to the opposite ear of the listener)
, And adds the filtered output with the filter coefficient set to Hl and the filtered output with the different filter coefficient Hr of the other channel based on the set filtered output, respectively, and filters the pair of added outputs. 13. The multi-channel signal processing method according to claim 8, wherein the processed signal is output as a pair of left and right front stereo processed signals. 14. The filter processing step, wherein the filter coefficient P is P = (F + K) / (S + A) (where S is a transfer function from a pair of speakers to an ear on the same side of the listener, and A is a pair of speakers , F is the transfer function from the position where the sound image is to be localized to the ear on the same side of the listener, and K is the transfer function from the position where the sound image is to be localized to the opposite ear of the listener. Transfer function up to)
And the first filter processing output obtained by filtering the sum signal of the pair of left and right front stereo signals and the filter coefficient N is set to N = (F−K) / (S−A). A sum signal and a difference signal of the first and second filtered outputs are obtained based on a second filtered output obtained by filtering the difference signal of the stereo signal, and the sum signal and the difference signal are filtered. The multi-channel signal processing method according to claim 8, wherein the signal is output as a pair of left and right front stereo processing signals.