JPH1188994A - Sound image presence device and sound image control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a sound image presence device with a smaller arithmetic operation amount than that of a conventional device and with about the same presence sense as that of the conventional device. SOLUTION: A sound signal is received from a signal source 4A, and a signal division means 6 divides the sound signal into L and R channel sound signals. A signal processing means 1 uses control speakers 8-1, 8-2 to process the input signal so as to provide an output of an imaginary sound image from a position of a virtual speaker 9. Thus, the transfer characteristic (frequency characteristic) of the signal processing means 1 is selected to be a characteristic equivalent to a difference of left right ears of a listener 10. Then its output is given to the L channel speaker 8-1 via a D/A converter 7-1. An output of the signal division means 6 is given to the R channel speaker 8-1 via a D/A converter 7-2. Thus, the R channel signal processing means is not required and number of components of FIR filters of the signal processing means is decreased and the amount of arithmetic operation is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sound image localization apparatus and a sound image control method for localizing a virtual sound image at an arbitrary position in an AV (audio / visual) device.

[0002]

2. Description of the Related Art In recent years, in the field of movies and broadcasting, digital sound compression technology has been used to record and reproduce multi-channels such as 5.1 channels. However, even if an attempt is made to reproduce a multi-channel audio signal in a general home, the sound output of a normal home television is often two or less. Therefore, it is expected that a multi-channel reproduction effect can be obtained even in an AV device having a two-channel sound reproduction function by using techniques such as sound field control and sound image control.

[0003] In recent years, techniques using frequency transformation such as MDCT (discrete cosine transformation) have been increasing as acoustic compression techniques. Here, in addition to the acoustic compression technique using frequency conversion, a conventional technique for performing sound image control will be described below.

FIG. 2 is a block diagram showing a basic configuration of a conventional sound image localization device (sound image reproduction device). First, a method of localizing a sound image to the front right of the listener 10 using the speakers 8-1 and 8-2 placed in front will be described. FIG.
As shown in the figure, the sound image localization apparatus includes a signal source 4, a signal dividing unit 6, signal processing units 1-1 and 1-2, and a D / A converter 7-.
1, 7-2 and control speakers 8-1 and 8-2.

The signal source 4 is signal input means for inputting a PCM audio signal S (t). The signal dividing means 6 is a means for distributing the audio signal S (t) to the left (L) channel and the right (R) channel. The signal processing means 1-1 has a transfer characteristic hL (n).
And a signal processing means 1-2.
Is a digital filter having a transfer characteristic hR (n). The digital output of the signal processing means 1-1 is supplied to a D / A converter 7-1.
Is converted to an analog audio signal by the control speaker 8 on the left side.
-1. Similarly, the digital output of the signal processing means 1-2 is converted into an analog audio signal by the D / A converter 7-2, and is provided to the right control speaker 8-2.

FIG. 3 is a block diagram showing the structure of the signal processing means 1-1 or 1-2. This signal processing means is an FIR filter composed of n-stage delay units (D) 11-1 to 11-n, n + 1 multipliers 12-1 to 12- (n + 1), and an adder 13. is there. Multipliers 12 are connected to the input / output terminals of the delay units 11 at the respective stages, and their outputs are added by an adder 13 and output.

The operation of the conventional sound image localization apparatus will be described with reference to FIGS. 2 and 3. In FIG. 2, the head-related transfer function between the speaker and the listener's ear is called an impulse response, and the value of the speaker 8-1 and the left ear of the listener 10 is h1 (t). Hereinafter, an impulse response will be used in the description in the time domain. This impulse response h1 (t) is exactly the response at the position of the eardrum of the left ear of the listener when an audio signal is input to the speaker 8-1, but when the measurement is performed, It shall be performed at the position. Note that similar results can be obtained in the frequency domain.

Similarly, let h 2 (t) be an impulse response between the speaker 8-1 and the right ear of the listener 10. Let h3 (t) be the impulse response between the speaker 8-2 and the left ear of the listener 10. Let h4 (t) be the impulse response between the speaker 8-2 and the right ear of the listener 10. It is assumed that the virtual speaker 9 is a virtual sound source localized to the front right of the listener. Let h5 (t) be the impulse response between the virtual speaker 9 and the left ear of the listener 10. h6 (t) is an impulse response between the virtual speaker 9 and the right ear of the listener 10.

In such a configuration, when the audio signal S (t) is emitted from the virtual speaker 9 from the signal source 4, the sound reaching the ear of the listener 10 is expressed by the following equation (1) and (2). It is represented by an expression. In the left ear, L (t) = S (t) * h5 (t) (1) In the right ear, R (t) = S (t) * h6 (t) (2) This represents a convolution operation. Actually, the transfer function of the speaker itself is multiplied, but is ignored here. Also, the transfer function of the speaker or the like is h5 (t), h5
6 (t).

The impulse response and the signal S (t) are
Time is considered as a discrete digital signal, and L (t) → L (n) R (t) → R (n) h5 (t) → h5 (n) h6 (t) → h6 (n) S (t ) → Notation like S (n). n represents an integer, and if T is a sampling time, n in parentheses is exactly nT. Here, T is omitted.

At this time, the expressions (1) and (2) are expressed as the following expressions (3) and (4), respectively, and the symbol * of the convolution operation is replaced by the multiplication symbol x. L (n) = S (n) × h5 (n) (3) R (n) = S (n) × h6 (n) (4)

Similarly, the signal S (t) is transmitted to the speaker 8-
The sound radiated from 1,8-2 and arriving at the left ear of the listener 10 is expressed by the following equation (5). L '(t) = S (t) * hL (t) * h1 (t) + S (t) * hR (t) * h3 (t) (5) The signal S (t) is transmitted to the speaker 8-1. , 8-2,
The sound that reaches the right ear of the listener 10 is expressed by the following equation (6). R '(t) = S (t) * hL (t) * h2 (t) + S (t) * hR (t) * h4 (t) (6)

When the expressions (5) and (6) are represented by using the impulse response with the expression (n), the following expressions (8) and (9) are obtained. L ′ (n) = S (n) × hL (n) × h1 (n) + S (n) × hR (n) × h3 (n) (8) R ′ (n) = S (n) × hL (n) × h2 (n) + S (n) × hR (n) × h4 (n) (9) where hL (n) is a transfer characteristic of the signal processing means 1-1,
hR (n) is a transfer characteristic of the signal processing means 1-2.

It is assumed that sound is heard from the same direction if the head related transfer functions are equal. This assumption is generally correct. Here, assuming the relationship of the following equation (10),
Equation (11) holds. L (n) = L ′ (n) (10) h5 (n) = hL (n) × h1 (n) + hR (n) × h3 (n) (11)

Similarly, assuming the relationship of equation (12),
Equation (13) holds. R (n) = R '(n) (12) h6 (n) = hL (n) * h2 (n) + hR (n) * h4 (n) (13)

In order for the listener 10 to be able to hear a predetermined sound from the front right of the virtual speaker 9 using the speakers 8-1 and 8-2, the following equation (1) is used.
The values of hL (n) and hR (n) may be determined so as to satisfy the expressions (1) and (13). For example, if Equations (11) and (13) are rewritten in the frequency domain expression, the convolution operation is replaced by multiplication, and the impulse response of each is converted to a transfer function by FFT. Since the transfer functions other than the FIR filter are obtained by measurement, the transfer function of the FIR filter can be obtained from these two equations.

HL (n) and hR (n) thus determined
From the speaker 8-1, the signal S (n) and hL (n) are radiated as a convolution, and the speaker 8-2 outputs the signal S (n).
By radiating the convolution of (n) and hR (n), the virtual speaker 9 at the right front can be actually emitted without being sounded.
The listener 10 can feel that the sound is sounding from the front right. The FIR filter shown in FIG. 3 can localize a sound image at an arbitrary position by the above signal processing.

[0018]

However, in the above configuration, in order to realize an actual head related transfer function,
It is necessary to provide an FIR filter for each channel and perform an extremely large number of convolution operations. In the FIR filter, when the filter order and the number of channels increase, the calculation speed and the load on hardware increase, which hinders realization. A method of reducing the number of taps of the FIR filter is also conceivable, but a certain number of taps is required in order to maintain the implementation accuracy of the head-related transfer function. When the number of taps is reduced, the sound image is blurred or the sound quality is deteriorated.

SUMMARY OF THE INVENTION The present invention has been made in view of such a conventional problem, and realizes a sound localization similar to a case where the number of taps of a digital filter is large with a small amount of calculation. It is an object of the present invention to provide a sound image localization device that can perform the above-mentioned operations and to realize a sound image control method thereof.

[0020]

In order to solve such a problem, the invention according to claim 1 of the present application provides a signal source for outputting an audio signal, and a signal source for outputting the audio signal from the signal source.
A signal dividing means for dividing the digital sound signal into two channels of digital sound signals, and one of the digital sound signals divided by the signal dividing means, and an imaginary sound image is localized at a predetermined position using a filter having a first frequency characteristic Signal processing first
, A first D / A converter for converting a digital signal output from the first signal processing means into an analog signal, and the other digital audio signal divided by the signal dividing means. , The second D /
An A converter, a first control speaker that radiates an audio signal converted by the first D / A converter to a predetermined space area, and an audio signal converted by the second D / A converter And a second control speaker that radiates the light into a predetermined space region.

According to a second aspect of the present invention, in the sound image localization apparatus according to the first aspect, the first frequency characteristic of the first signal processing means is set so that a listener can receive the first frequency characteristic from the first and second control speakers. Are determined so as to realize a binaural difference between sounds reaching the left and right ears of the listener from the virtual sound image.

According to a third aspect of the present invention, there is provided a signal source for outputting an audio signal, and a second signal for subjecting the audio signal output from the signal source to signal processing using a filter having a second frequency characteristic. Processing means; a signal dividing means for dividing the sound signal output from the second signal processing means into two-channel digital sound signals; and one of the digital sound signals divided by the signal dividing means, A first signal processing means for performing signal processing so that a virtual sound image is localized at a predetermined position by using a filter having a frequency characteristic of, and a first signal for converting a digital signal output from the first signal processing means into an analog signal. A D / A converter, a second D / A converter that receives the other digital audio signal divided by the signal dividing unit and converts the signal into an analog signal, and a conversion by the first D / A converter Sound signal And a second control speaker that radiates the acoustic signal converted by the second D / A converter to a predetermined space region. It is a feature.

According to a fourth aspect of the present invention, in the sound image localization apparatus according to the third aspect, the first frequency characteristic of the first signal processing means is set such that the first frequency characteristic of the listener is controlled by the first and second control speakers. Are determined so as to realize the binaural difference between the sounds reaching the right and left ears of the listener from the virtual sound image from the virtual sound image. The frequency characteristic is characterized by correcting at least one of a sound quality, a volume change, and a phase characteristic of the first frequency characteristic of the first signal processing means.

The invention according to claim 5 of the present application uses a signal source for outputting a sound signal in the frequency domain and a sound signal in the frequency domain output from the signal source using a filter having a third frequency characteristic. Signal processing means for performing signal processing on the frequency domain, frequency-time domain conversion means for converting a frequency-domain audio signal output from the third signal processing means into a time-domain audio signal, and the frequency-time domain A signal dividing unit that divides an output signal of the converting unit into two-channel digital acoustic signals, and one of the digital acoustic signals divided by the signal dividing unit is input to a predetermined position using a filter having a first frequency characteristic. A first signal processing means for performing signal processing so that a virtual sound image is localized, and a first D for converting a digital signal output from the first signal processing means into an analog signal.
/ A converter, and a second digital audio signal which is input by the other digital audio signal divided by the signal dividing means and is converted into an analog signal.
D / A converter, a first control speaker that radiates an acoustic signal converted by the first D / A converter to a predetermined space area, and a D / A converter that is converted by the second D / A converter. And a second control speaker that emits the generated acoustic signal to a predetermined spatial region.

According to a sixth aspect of the present invention, in the sound image localization apparatus of the fifth aspect, the first frequency characteristic of the first signal processing means is set so that the first frequency characteristic of the listener is controlled by the first and second control speakers. Are determined so as to realize the binaural difference between the sounds reaching the left and right ears of the listener from the virtual sound image from the virtual sound image, respectively. The frequency characteristic is characterized in that at least one of sound quality, volume change and phase characteristic of the first frequency characteristic of the first signal processing means is corrected in a frequency domain.

According to a seventh aspect of the present invention, a signal source for outputting an audio signal in a frequency domain and an audio signal in a frequency domain output from the signal source are used by using a filter having a third frequency characteristic. Signal processing means for performing signal processing, frequency-time domain conversion means for converting a frequency-domain signal output from the third signal processing means into a time-domain signal, and the frequency-time domain conversion means Signal processing means for performing signal processing on the output sound signal using a filter having a second frequency characteristic, and signal division for dividing the output signal of the second signal processing means into two-channel digital sound signals Means for inputting one digital audio signal divided by the signal dividing means, and performing signal processing using a filter having a first frequency characteristic so as to localize a virtual sound image at a predetermined position. Processing means, a first D / A converter for converting a digital signal output from the first signal processing means into an analog signal, and the other digital audio signal divided by the signal dividing means, A second D / A converter for converting the signal into a signal, a first control speaker for radiating the acoustic signal converted by the first D / A converter to a predetermined spatial area, and a second D / A converter. And a second control speaker for radiating the acoustic signal converted by the A converter to a predetermined space region.

According to an eighth aspect of the present invention, in the sound image localization apparatus according to the seventh aspect, the first frequency characteristic of the first signal processing means is set such that the first frequency characteristic of the listener is controlled by the first and second control speakers. Are determined so as to realize the binaural difference between the sounds reaching the left and right ears of the listener from the virtual sound image from the virtual sound image, respectively. The synthesized frequency characteristic of the frequency characteristic and the second frequency characteristic in the second signal processing means is obtained by converting at least one of the sound quality, volume change, and phase characteristic of the first frequency characteristic of the first signal processing means into a frequency domain. The above correction is performed.

According to a ninth aspect of the present invention, a sound image is provided at an arbitrary position as viewed from the listener by using the first control speaker and the second control speaker arranged in the left and right spaces of the listener. Is a sound image control method in which signal processing means is provided only for a signal input to a first control speaker to perform a virtual sound image localization, wherein the left and right sides of a listener from the first and second control speakers are The sound arriving at the ears is determined from the virtual sound image a frequency characteristic G (n) that realizes a binaural difference between the sounds arriving at the left and right ears of the listener, and G (n) is calculated as the frequency characteristic of the signal processing means. Thus, the acoustic signal of the signal source is localized at the position of the virtual sound image.

According to a tenth aspect of the present invention, in the sound image control method of the ninth aspect, the frequency characteristic G (n)
Implies the impulse responses of the first control speaker and the left and right ears of the listener as h1 (t) and h2 (t), and the impulse responses of the second control speaker and the left and right ears of the listener. H3
(t) and h4 (t), and when a virtual sound source localized in an arbitrary direction of the listener is a virtual speaker, impulse responses between the virtual speaker and the left and right ears of the listener are h5 (t) and h6.
When (t) is used, (1) when an acoustic signal S (t) is emitted from a virtual speaker from a signal source, the sound reaching the left ear of the listener is L (t) = S (t) * h5 (t) ), And the sound reaching the right ear of the listener is calculated as R (t) = S (t) * h6 (t). (2) The signals L (t), R (t), h5 ( t), h6 (t), S (t)
To the discrete signals L (n), R (n), h5 (n), h6
(n), S (n), and (3) L (n) = S (n) × h5
(n) and R (n) = S (n) × h6 (n) are obtained.
L '(t) = S (t) * hL (t) * h1 (t) + S (t) * h
(5) The sound radiated from the second control speaker and reaching the right ear of the listener is calculated as R '(t).
= S (t) * hL (t) * h2 (t) + S (t) * hR (t) * h4 (t)
(6) L ′ (t) is calculated as L ′ (n) = S (n) × h
L (n) × h1 (n) + S (n) × hR (n) × h3 (n)
(7) R ′ (t) is calculated as follows: R ′ (n) = S (n) × hL (n) × h2 (n) +
S (n) × hR (n) × h4 (n), and (8) L (n) =
Assuming L '(n), h5 (n) = hL (n) * h1 (n) + h
R (n) × h3 (n), (9) Assuming that R (n) = R ′ (n), h6 (n) = hL (n) × h2 (n) + hR (n) × h4 ( n) and (10) hL (n) and hR (n) are calculated from the above (8) and (9), and G (n) = hL (n) / hR (n). It is characterized by the following.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 A sound image localization apparatus according to Embodiment 1 of the present invention will be described with reference to the drawings. FIG.
FIG. 2 is a block diagram showing the overall configuration of the sound image localization apparatus according to the present embodiment, in which the same parts as those in the conventional example are denoted by the same reference numerals, and detailed description will be omitted. In the sound image localization apparatus shown in FIG. 1, the first signal processing means 1 is provided only on one channel (here, the left channel). The signal processing means 1 is composed of the FIR filter shown in FIG.

In FIG. 1, when a digitally converted (PCM) audio signal S (t) is inputted from the signal source 4A, the signal dividing means 6 divides this signal into two signals. One of the divided signals is a D / A converter 7-2.
And the other is input to the signal processing means 1.

In the conventional example, if the frequency characteristics of the signal processing means 1-1 and 1-2 for realizing the function of the virtual speaker 9 are Hl (n) and HR (n), the signal processing means of the present embodiment 1 is set to the frequency characteristic of Hl (n) -HR (n). That is, the frequency characteristic of the binaural difference of the listener is set by the signal processing means 1 as a transfer function. In the present embodiment, the frequency characteristic of the signal processing unit 1 is obtained by using division in the frequency domain. However, the frequency characteristic can be similarly obtained by using various methods such as deconvolution, and in any case, Similar effects can be obtained.

The frequency characteristic G (n) of the signal processing means 1 can be obtained by the following equation (14). G (n) = hL (n) / hR (n) (14) Here, the frequency characteristic G (n) of the signal processing means 1 is the frequency characteristic hL (n) of the signal processing means 1-1 in the conventional example. ) And the frequency characteristic hR (n) of the signal processing means 1-2.

FIG. 4 shows an example of the frequency characteristics of the signal processing means 1-1, and FIG. 6 shows an example of the frequency characteristics of the signal processing means 1-2. FIG. 8 shows an example of the frequency characteristics of the signal processing means 1 in this embodiment in this case. In order to avoid the overflow of the calculation result, as the frequency characteristics of the signal processing means 1, of the frequency characteristics in FIG. 4 and FIG. 6, the division with the higher sound pressure level as the denominator and the lower sound pressure level as the numerator. I do. In the case of using an arithmetic method that does not overflow, either the numerator or the denominator of the division may be either value.

FIGS. 5 and 5 show examples of tap coefficients of the FIR filter having the frequency characteristics shown in FIGS.
FIG. 7 and FIG. FIG. 5 is an example of coefficients of an FIR filter constituting the signal processing means 1-1. FIG. 7 is an example of coefficients of an FIR filter constituting the signal processing means 1-2. FIG.
Is a coefficient example of the FIR filter constituting the signal processing means 1. The signal processing means 1-1 and 1-2 of the conventional example each required about 128 taps (total of 256 taps), whereas the signal processing means 1 of the present embodiment comprises a filter of about 128 taps. can do. The output signal of the signal dividing means 6 is given to the signal processing means 1 having such a transfer characteristic.

In the conventional example, if the sound reaching the left ear of the listener 10 is YL (n) and the sound reaching the right ear is YR (n), the following equations (15) and (16) hold. YL (n) = S (n) .times.hL (n) .times.h1 (n) + S (n) .times.hR (n) .times.h3 (n)... (15) YL (n) = S (n) .times.hL (n) × h2 (n) + S (n) × hR (n) × h4 (n) (16)

In the present embodiment, assuming that the sound reaching the left ear of the listener 10 is Y'L (n) and the sound reaching the right ear is Y'R (n), the following equation (17): 18) The equation holds. Y′L (n) = S (n) × hL (n) / hR (n) × h1 (n) + S (n) × h3 (n) (17) Y′R (n) = S ( n) × hL (n) / hR (n) × h2 (n) + S (n) × h4 (n) (18) When this equation is compared with the equations (15) and (16), the conventional example is obtained. It can be seen that the input signal has the same transfer characteristic as 1 / hR (n) (19).

As a result, the sound quality of the output signal becomes the one obtained by filtering the transfer characteristic of (19) with respect to the conventional example, and the treble range tends to be enhanced. However, since the sense of localization does not change, this method is a very effective method when sound quality is not emphasized.

As described above, it is possible to obtain almost the same sense of localization as in the conventional method with a smaller amount of calculation than in the conventional method, and to emit sound to a predetermined area using the speakers 8-1 and 8-2, thereby realizing a virtual reality. An effect as if a sound image was emitted from the virtual speaker 9 can be obtained.

(Embodiment 2) Hereinafter, a sound image localization apparatus according to Embodiment 2 of the present invention will be described with reference to the drawings. FIG. 10 is a block diagram showing the overall configuration of the sound image localization apparatus according to the present embodiment. The same blocks as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted.
As shown in the figure, the second signal processing means 2 is provided between the signal source 4A and the signal dividing means 6. The signal processing means 2 is constituted by an FIR filter or IIR filter as shown in FIG. The frequency characteristics of the signal processing means 2 will be described later.

The signal processed by the signal processing means 2 is divided by the signal dividing means 6 into two signals. One is D /
The signal is input to the A converter 7-2, and the other is input to the first signal processing means 1. The signal processing means 1 has an F
It is constituted by an IR filter.

Frequency characteristics (transfer function) of signal processing means 1
Are set to have the same frequency characteristics as in the first embodiment. Here also, the frequency characteristic of the signal processing means 1 is obtained using division in the frequency domain. Further, the frequency characteristics can be similarly obtained using various methods such as deconvolution. The same effect can be obtained by using either method.

Frequency characteristics (transfer characteristics) of signal processing means 2
When the frequency characteristic of the signal processing means 1 is obtained, the frequency characteristic is a denominator, that is, the frequency characteristic of hR (n). This can be expressed by the following equation (17), (18)
Equation (20) is obtained. Y′L (n) = S ′ (n) × hL (n) / hR (n) × h1 (n) + S ′ (n) × h3 (n) (17) Y′R (n) = S ′ (n) × hL (n) / hR (n) × h2 (n) + S ′ (n) × h4 (n) (18) S ′ (n) = S (n) × hR (n ) ・ ・ ・ (20)

To summarize this, the following equation (21), (2)
2) It is expressed like the equation. Y′L (n) = S (n) × hL (n) × h1 (n) + S (n) × hR (n) × h3 (n) (21) Y′R (n) = S ( n) × hL (n) × h2 (n) + S (n) × hR (n) × h4 (n) (22) This is consistent with equations (14) and (15). As a result, in the present embodiment, the sound quality changes little and the sound source can be localized at the position of the virtual sound image in the transfer characteristics similar to those of the conventional example, that is, in any of the low sound, the medium sound, and the high sound.

Since the sense of localization of the sound image can be realized by the signal processing means 1, the signal processing means 2 can be realized with an order smaller than that of the conventional filter. In this embodiment, a 32-tap filter is used as the signal processing means 2 to realize a sound image localization apparatus having the same sound quality as that of the related art.

FIG. 11 shows an example of the frequency characteristic of the FIR filter used in the signal processing means 2, and FIG. 12 shows the tap coefficients. As a result, the amount of computation was: Conventional example: 128 tap convolution twice This embodiment: 128 tap convolution + 32 tap convolution As a result, it is possible to obtain almost the same sound quality and sense of localization as in the related art with a calculation amount of ／ of the conventional example.

The output signal of the signal processing means 1 is input to a D / A converter 7-1, and the other output signal of the signal dividing means 6 is input to a D / A converter 7-2. Then, the outputs of the D / A converters 7-1 and 7-2 are given to the speakers 8-1 and 8-2, respectively. Then, by emitting sound to a predetermined area, the virtual sound image can be localized at the position of the virtual speaker 9.

(Embodiment 3) Next, a sound image localization apparatus according to Embodiment 3 of the present invention will be described with reference to the drawings. FIG. 13 is a block diagram showing the overall configuration of the sound image localization apparatus according to the present embodiment. The same blocks as those in Embodiment 2 are given the same names, and detailed descriptions are omitted.
In the sound image localization apparatus according to the present embodiment, the signal source 4B outputs a sound signal in the frequency domain.
(Aadptive Transform Acoustic Coding) MD
The signal of the component of CT is output. The third signal processing means 3 and the frequency-time domain converter 5 are provided between the signal source 4B and the signal dividing means 6.

The structure of the first signal processing means 1 is the same as that shown in FIG. FIG. 14 shows a configuration example of the third signal processing means 3. The signal processing means 3 receives the input signal (X1,
X2,... Xn) are multiplied by coefficients (a1, a2,.
Multipliers 12-1, 12-2,..., 12-amplifying in n)
n, and their output signals are (Y1, Y2,.
n).

In FIG. 13, the digital audio signal converted into each frequency component from the signal source 4B is input to the signal processing means 3. The signal processing means 3 processes the input signal for each frequency band as shown in FIG. 14 in order to localize the sound image. That is, signals of each frequency band (X1, X2,...)
Xn), the multiplication coefficients (a1, a2,...)
a). The amplification factor a of each multiplier 12 is determined by the frequency-time conversion method in the frequency-time converter 5.

The output signal of the signal processing means 3 is input to the frequency-time converter 5 and is converted into a time-domain signal (PCM sound signal). The signal dividing means 6 divides the input signal into two signals, one of which is supplied to the D / A converter 7-2, and the other is supplied to the first signal processing means 1.

Frequency characteristics (transfer function) of signal processing means 1
Are set in the same manner as in the first embodiment. In the present embodiment, the signal processing means 1 uses division in the frequency domain.
, The frequency characteristic can be similarly obtained by using various methods such as deconvolution.
In each case, a similar effect can be obtained.

Frequency characteristics (transfer characteristics) of signal processing means 3
Is preferably the frequency characteristic that is the denominator of the equation (14), that is, the frequency characteristic of hR (n). This can be expressed by the following equations (23), (24) and (25). Y′L (n) = S ′ (n) × hL (n) / hR (n) × h1 (n) + S ′ (n) × h3 (n) (23) Y′R (n) = S ′ (n) × hL (n) / hR (n) × h2 (n) + S ′ (n) × h4 (n) (24) S ′ (n) = S (n) × hR (n ) ・ ・ ・ (25)

When this is arranged, Equation (26) and (27)
It looks like an expression. Y′L (n) = S (n) × hL (n) × h1 (n) + S (n) × hR (n) × h3 (n) (26) Y′R (n) = S ( n) × hL (n) × h2 (n) + S (n) × hR (n) × h4 (n) (27) This is consistent with the equations (15) and (16). Thus, it can be seen that the present embodiment can achieve the function of the sound image localization device having the same transfer characteristics as the conventional example.

On the other hand, since the signal processing means 3 can be realized by a plurality of multipliers, the amount of calculation is determined by the conversion length of the frequency-time converter. Here, when the conversion length is m, the amount of calculation is: Conventional example: 128 tap convolution twice = 128 × m × 2 = 2
56m Embodiment: 128-tap convolution + m-order multiplier = 1
28 × m + m = 129 m. Therefore, the sound quality and the sense of localization that are substantially the same as those of the conventional example can be obtained with about 1/2 the amount of calculation of the conventional example.

Next, the output signal of the signal processing means 1 is supplied to a D / A converter 7-1, and the output signal of the signal dividing means 6 is converted to a D / A signal.
It is given to the converter 7-2. Then, the D / A converter 7-1,
The output of 7-2 is given to speakers 8-1 and 8-2, respectively, and a sound is emitted to a predetermined area. Thus, the virtual sound image can be localized at the position of the virtual speaker 9.

(Embodiment 4) Next, a sound image localization apparatus according to Embodiment 4 of the present invention will be described with reference to the drawings. FIG. 15 is a block diagram showing the overall configuration of the sound image localization apparatus according to the present embodiment. The same blocks as those in Embodiments 1 to 3 are denoted by the same reference numerals, and detailed description is omitted. Here, a third signal processing means 3, a frequency-time domain converter 5, and a second signal processing means 2 are provided between the signal source 4B and the signal dividing means 6.

The structures of the first signal processing means 1 and the second signal processing means 2 are the same as those shown in FIG. Also the third
The structure of the signal processing means 3 is the same as that shown in FIG.

In FIG. 15, a digital audio signal converted into a frequency signal from the signal source 4B is input to the signal processing means 3. The signal processing means 3 processes the input signal with a predetermined multiplication coefficient for each frequency band in order to localize the sound image. The multiplication coefficient of the multiplier 12 is determined by the frequency-time conversion method in the frequency-time domain converter 5.

When the output signal of the signal processing means 3 is input to the frequency-time domain converter 5, it is converted into a time domain signal and input to the signal processing means 2. The signal processing means 2 converts the frequency characteristic using an FIR filter having a small number of taps, and corrects at least one of a phase component, a sound volume, and a sound quality which could not be corrected by the signal processing means 3 on a time axis. is there.

The signal processed by the signal processing means 2 is divided by the signal dividing means 6 into two signals. One of the divided signals is output to the D / A converter 7-2, and the other is output to the signal processing means 1.

The frequency characteristic of the signal processing means 1 can be obtained by the equation (14) as in the first embodiment. The product of the frequency characteristics (transfer characteristics) of the signal processing means 2 and 3 is
When the frequency characteristic of the signal processing means 1 is obtained in the conventional example, it is preferable that the frequency characteristic be a denominator, that is, the frequency characteristic of hR (n). This can be expressed by the following equations (23), (24), and (25). Y′L (n) = S ′ (n) × hL (n) / hR (n) × h1 (n) + S ′ (n) × h3 (n) (23) Y′R (n) = S ′ (n) × hL (n) / hR (n) × h2 (n) + S ′ (n) × h4 (n) (24) S ′ (n) = S (n) × hR (n ) ・ ・ ・ (25)

When this is arranged, Equation (26) and (27)
An expression is obtained. Y′L (n) = S (n) × hL (n) × h1 (n) + S (n) × hR (n) × h3 (n) (26) Y′R (n) = S ( n) × hL (n) × h2 (n) + S (n) × hR (n) × h4 (n) (27) This is consistent with the equations (15) and (16). Thus, it is understood that the present embodiment achieves the function of the sound image localization device having the same transfer characteristics as the conventional example.

Since the signal processing means 3 can be realized by a plurality of multipliers, the amount of operation is determined by the conversion length of the frequency-time converter. Here, when the conversion length is m, the calculation amount is the conventional example: 128 tap convolution twice = 128 × mx × 2 = 2
56m Embodiment: 128-tap convolution + m-order multiplier = 1
28 × m + m = 129 m. Therefore, the sound quality and the sense of localization that are almost the same as those of the related art can be obtained with about 1/2 the amount of calculation of the related art.

Next, the output signal of the signal processing means 1 is supplied to a D / A converter 7-1, and the output signal of the signal dividing means 6 is converted to a D / A signal.
It is given to the converter 7-2. Then, the D / A converter 7-1,
7-2 is output to speakers 8-1 and 8-2, respectively.
Emits sound to a given area. Thus, the virtual sound image can be localized at the position of the virtual speaker 9.

The frequency characteristic of the signal processing means 1 can be obtained by using division in the frequency domain, but can also be obtained by using various methods such as deconvolution.

[0067]

As described above, the sound image localization apparatus according to the first to eighth aspects of the present invention can realize a sound image localization apparatus having a smaller amount of calculation than that of the conventional method and having the same level of localization feeling.

In particular, according to the first and second aspects of the present invention, the signal processing conventionally performed on each channel is performed on only one channel, thereby realizing a sound image localization apparatus having less arithmetic processing than the conventional apparatus.

In particular, in the inventions of claims 3 and 4 of the present application, the signal processing conventionally performed on each channel is performed on only one channel and the sound quality is adjusted, so that the arithmetic processing is less than that of the conventional apparatus. A sound image localization device having the same sound quality as that of the related art is realized.

In particular, according to the fifth and sixth aspects of the present invention, the signal processing, which has been conventionally performed on each channel, is performed on only one channel, and the sound quality is adjusted in the frequency domain. And a sound image localization apparatus having the same sound quality as that of the related art is realized.

In particular, in the inventions of claims 7 and 8 of the present application, signal processing conventionally performed on each channel is performed on only one channel, and only sound quality adjustment is performed on the frequency domain and the time domain. A sound image localization device having less arithmetic processing than the device and having the same sound quality as the conventional device is realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of a sound image localization apparatus according to Embodiment 1 of the present invention.

FIG. 2 is a block diagram illustrating a configuration example of a conventional sound image localization device.

FIG. 3 is a configuration diagram of an FIR filter used for a signal processing unit in each embodiment of the present invention.

FIG. 4 is a frequency characteristic diagram of an Lch signal processing unit in a conventional sound image localization apparatus.

FIG. 5 is an explanatory diagram showing filter coefficients (time characteristics) of Lch signal processing means in a conventional sound image localization apparatus.

FIG. 6 is a frequency characteristic diagram of an Rch signal processing unit in a conventional sound image localization apparatus.

FIG. 7 is an explanatory diagram showing filter coefficients (time characteristics) of an Rch signal processing unit in a conventional sound image localization apparatus.

FIG. 8 is a frequency characteristic diagram of the signal processing means in the first embodiment.

FIG. 9 is an explanatory diagram showing filter coefficients (time characteristics) of the signal processing means according to the first embodiment.

FIG. 10 is a block diagram of a sound image localization apparatus according to Embodiment 2 of the present invention.

FIG. 11 is a frequency characteristic diagram of the signal processing means 2 according to the second embodiment.

FIG. 12 is an explanatory diagram showing filter coefficients (time characteristics) of the signal processing means 2 according to the second embodiment.

FIG. 13 is a block diagram of a sound image localization apparatus according to Embodiment 3 of the present invention.

FIG. 14 is a configuration diagram of a signal processing unit 3 according to the third embodiment.

FIG. 15 is a block diagram of a sound image localization apparatus according to Embodiment 4 of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 first signal processing means 2 second signal processing means 3 third signal processing means 4 signal source 5 frequency-time domain converter 6 signal dividing means 7-1, 7-2 D / A converter 8-1 , 8-2 speaker 9 virtual speaker 10 listener 11-1 to 11-n delay unit 12-1 to 12- (n + 1) multiplier 13 adder

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Masahiro Sueyoshi 1006 Kazuma Kadoma, Osaka Prefecture Inside Matsushita Electric Industrial Co., Ltd. 72) Inventor Shuji Miyasaka 1006 Kazuma Kadoma, Kadoma City, Osaka Prefecture, Japan Inside Matsushita Electric Industrial Co., Ltd. 1006 Kadoma Kadoma Matsushita Electric Industrial Co., Ltd. (72) Inventor Sou Ishido 1006 Kadoma Kadoma, Osaka Pref.Matsushita Electric Industrial Co., Ltd. Denki Sangyo Co., Ltd.

Claims (10)

[Claims] 1. A signal source for outputting an audio signal, a signal dividing unit for dividing an audio signal output from the signal source into digital audio signals of two channels, and one digital signal divided by the signal dividing unit A first signal processing unit that receives an acoustic signal and performs signal processing so that a virtual sound image is localized at a predetermined position using a filter having a first frequency characteristic; and a digital signal output from the first signal processing unit. A first D / A converter that converts the signal into an analog signal; a second D / A converter that receives the other digital sound signal divided by the signal dividing unit and converts the signal into an analog signal; A first control speaker that radiates an acoustic signal converted by the D / A converter to a predetermined space area; and radiates an audio signal converted by the second D / A converter to a predetermined space area. Second control speed A sound image localization device comprising: 2. A first frequency characteristic of the first signal processing means, wherein sounds reaching the left and right ears of the listener from the first and second control speakers respectively are different from the listener's right image. 2. The sound image localization device according to claim 1, wherein the sound image localization device is determined so as to realize a binaural difference between sounds reaching left and right ears. 3. A signal source for outputting an audio signal, second signal processing means for performing signal processing on the audio signal output from the signal source using a filter having a second frequency characteristic, and the second signal A signal dividing unit that divides the acoustic signal output from the processing unit into two-channel digital acoustic signals; and one of the digital acoustic signals divided by the signal dividing unit is input, and a predetermined signal is input using a filter having a first frequency characteristic. A first signal processing unit that performs signal processing so that a virtual sound image is localized at the position of the first signal processing unit; a first D / A converter that converts a digital signal output from the first signal processing unit into an analog signal; A second D / A converter that receives the other digital audio signal divided by the division unit and converts the digital audio signal into an analog signal; and converts the audio signal converted by the first D / A converter into a predetermined spatial domain. Radiate to A sound image localization device comprising: a first control speaker; and a second control speaker that emits an acoustic signal converted by the second D / A converter to a predetermined spatial region. 4. A first frequency characteristic of the first signal processing means, wherein sounds reaching the right and left ears of the listener from the first and second control speakers respectively are different from the listener's right image. The second frequency characteristic of the second signal processing means is determined so as to realize a binaural difference between sounds reaching left and right ears. 4. The sound image localization apparatus according to claim 3, wherein at least one of sound quality, volume change, and phase characteristics is corrected. 5. A signal source for outputting a sound signal in the frequency domain, and a third signal processing for processing the sound signal in the frequency domain output from the signal source using a filter having a third frequency characteristic. Means, frequency-time domain conversion means for converting a frequency domain audio signal output from the third signal processing means into a time domain audio signal, and a two-channel output signal of the frequency-time domain conversion means. Signal dividing means for dividing the digital sound signal into digital sound signals, and one of the digital sound signals divided by the signal dividing means is inputted, and signal processing is performed using a filter having a first frequency characteristic so that a virtual sound image is localized at a predetermined position. A first D / A converter for converting a digital signal output from the first signal processing unit into an analog signal; A second D / A converter that receives the digital audio signal and converts the digital audio signal into an analog signal; and a first D / A converter that radiates the audio signal converted by the first D / A converter to a predetermined spatial region. A sound image localization device comprising: a control speaker; and a second control speaker that radiates an acoustic signal converted by the second D / A converter to a predetermined space region. 6. A first frequency characteristic of the first signal processing means, wherein sounds reaching the right and left ears of the listener from the first and second control speakers respectively are different from the listener's right-hand ears. The third frequency characteristic of the third signal processing means is determined so as to realize a binaural difference between sounds reaching left and right ears. 6. The sound image localization apparatus according to claim 5, wherein at least one of sound quality, volume change, and phase characteristics is corrected in a frequency domain. 7. A signal source for outputting a sound signal in a frequency domain, and a third signal processing for processing the sound signal in a frequency domain output from the signal source using a filter having a third frequency characteristic. Means, frequency-time domain conversion means for converting a frequency domain signal output by the third signal processing means into a time domain signal, and an output sound signal of the frequency-time domain conversion means,
A second signal processing unit that performs signal processing using a filter having the following frequency characteristics: a signal dividing unit that divides an output signal of the second signal processing unit into a two-channel digital acoustic signal; A first signal processing unit that receives one of the divided digital audio signals and performs signal processing using a filter having a first frequency characteristic so that a virtual sound image is localized at a predetermined position; and the first signal processing unit. A first D / A converter for converting a digital signal output from the digital signal into an analog signal, and a second D / A converter for receiving the other digital audio signal divided by the signal dividing means and converting the signal into an analog signal A first control speaker that radiates the sound signal converted by the first D / A converter to a predetermined space region; and outputs a predetermined sound signal converted by the second D / A converter to a predetermined space area. Spatial domain of Sound image localization apparatus characterized by comprising a second control speaker for radiating a. 8. A first frequency characteristic of the first signal processing means, wherein sounds reaching the left and right ears of the listener from the first and second control speakers respectively are different from the listener's right-hand ear from the virtual sound image. Determining a binaural difference between sounds reaching the left and right ears, and combining a third frequency characteristic in the third signal processing means and a second frequency characteristic in the second signal processing means. 8. The sound image according to claim 7, wherein the frequency characteristic corrects at least one of a sound quality, a volume change, and a phase characteristic of the first frequency characteristic of the first signal processing means in a frequency domain. Localization device. 9. A first control when a sound image is localized at an arbitrary position viewed from a listener using a first control speaker and a second control speaker arranged in a space on the left and right sides of the listener. A sound image control method in which signal processing means is provided only for a signal input to a speaker to perform a virtual sound image localization, wherein sounds reaching the right and left ears of a listener from the first and second control speakers, respectively, are: A frequency characteristic G (n) for realizing a binaural difference between sounds reaching the right and left ears of the listener from the virtual sound image is obtained, and G (n)
Is a frequency characteristic of the signal processing means, thereby localizing the acoustic signal of the signal source at the position of the virtual sound image. 10. The frequency characteristic G (n) is such that the impulse responses between the first control speaker and the left and right ears of the listener are h1 (t) and h2 (t), and the second control speaker When the impulse responses of the listener's left and right ears are h3 (t) and h4 (t), and the virtual sound source localized in any direction of the listener is a virtual speaker, the virtual speaker and the listener's left ear When the impulse response with the right ear and the right ear is h5 (t) and h6 (t), (1) when the acoustic signal S (t) is emitted from the virtual speaker from the signal source, the sound reaching the left ear of the listener Is L (t) = S
(t) * h5 (t), and the sound reaching the right ear of the listener is R (t)
= S (t) * h6 (t), (2) Signals L (t), R (t), h5 (t), h6 on the time axis
(t) and S (t) are represented by discrete signals L (n), R (n) and h5, respectively.
(n), h6 (n), and S (n). (3) L (n) = S (n) × h5 (n) and R (n) = S (n) ×
h4 (n) is obtained. (4) The sound radiated from the first control speaker and reaching the listener's left ear is represented by L '(t) = S (t) * hL (t) * h1 (t) + S (t) * hR (t) *
(5) The sound radiated from the second control speaker and arriving at the right ear of the listener is represented by R '(t) = S (t) * hL (t) * h2 (t ) + S (t) * hR (t) *
(6) L ′ (t) is calculated as L ′ (n) = S (n) × hL (n) × h1 (n) +
S (n) × hR (n) × h3 (n), and (7) R ′ (t) is converted to R ′ (n) = S (n) × hL (n) × h2 (n) +
S (n) × hR (n) × h4 (n), (8) h5 (n) = hL (n) × h1 (n) in order to assume that L (n) = L ′ (n) + HR (n) × h3 (n). (9) Assuming that R (n) = R ′ (n), h6 (n) = hL (n) × h2 (n) + hR (n) × h4 ( (10) hL (n) and hR (n) are calculated from the above (8) and (9), and G (n) = hL (n) / hR (n). 10. The sound image control method according to claim 9, wherein: