WO2006126473A1

WO2006126473A1 - Sound image localization device

Info

Publication number: WO2006126473A1
Application number: PCT/JP2006/310120
Authority: WO
Inventors: Kazue Fusakawa
Original assignee: Matsushita Electric Industrial Co., Ltd.
Priority date: 2005-05-23
Filing date: 2006-05-22
Publication date: 2006-11-30

Abstract

[PROBLEMS] To provide a sound image localizing device capable of reproducing a natural sound feel without deteriorating the localized feel of a center localizing signal component, as might otherwise often occur in a speech or vocal sound localized at the center when outputted by an ordinary stereophonic speaker. [MEANS FOR SOLVING THE PROBLEMS] The sound localizing device comprises L-channel and R-channel sound image localization processing means (1) and (2) for subjecting stereophonic input acoustic signals of L-channel and R-channel to individual sound image localization processing treatments. Further comprised is a vocal band detection unit (14) for detecting the left and right bands of substantially identical phases as vocal signal bands from the individual input acoustic signals of the L-channel and the R-channel. Still further comprised is a mixing processing unit (11) for extracting the vocal signal components from the individual input acoustic signals of the L-channel and the R-channel on the basis of the bands detected by the vocal band detection unit (14), and adding the extracted vocal signal components to the sound image localization processing signals having been subjected to the sound image localization treatments.

Description

Specification

Sound image localization device

Technical field

[0001] The present invention relates to a three-dimensional sound image for localizing a stereo signal output from the stereo speaker at an interval wider than an actual speaker interval in a stereo speaker arranged adjacent to the front of the listener. It relates to a localization device.

Background art

[0002] As a conventional sound image localization device, in order to localize a sound image in a desired direction, an input stereo acoustic signal force, a first filter that removes the influence of a transfer function reaching the left and right ears, and an input stereo sound A second filter that gives a transfer characteristic in a desired direction to the signal, and a plurality of speakers that are arranged at predetermined positions and that output stereo sound signals localized by the first filter and the second filter. Some stereo speakers are configured (for example, refer to Patent Document 1). In this conventional sound image localization device, when the distance between the arranged speakers is relatively narrow and a stereo image is used to localize the sound image at a position wider than the actual speaker distance, the left and right input acoustic signals are used. This was realized by performing 3D sound image localization filtering according to the opening angle of the virtual speaker.

Patent Document 1: JP-A-10-70797 (Page 3-5, Fig. 1)

Disclosure of the invention

Problems to be solved by the invention

[0003] However, in the conventional sound image localization apparatus, when localization processing is performed on the left and right stereo input sound signals, if the sound signal components of speech and vocal sounds are high, the sound is output to the center. For a signal with the same phase on the left and right (hereinafter referred to as “center localization signal”), the sound wave is likely to be out of phase by the sound image localization process. Therefore, there was a problem that the distance of the sound heard from the center direction was felt far away.

[0004] The present invention solves the above-described conventional problems, and is common in speech and vocal sounds. An object of the present invention is to provide a sound image localization device that can reproduce a natural surround feeling without impairing the localization feeling of the sound component localized at the center when output from a stereo speaker.

Means for solving the problem

[0005] The sound image localization apparatus of the present invention performs sound image localization processing on an input stereo sound signal and outputs sound image localization processing signals, and is substantially the same on the left and right from the input stereo sound signals. Center localization signal detection means for detecting the center localization signal component band of the phase as the center localization signal band, and the center localization signal component based on the center localization signal band detected by the center localization signal component detection means It has a configuration comprising mixing means for extracting from the force stereo signal and adding it to the sound image localization processing signal output by the sound image localization processing means.

[0006] With this configuration, the sound image localization apparatus of the present invention reproduces the natural surround sound that does not impair the sense of localization of the sound component that is localized at the center when output from a normal stereo speaker, which is common in speech and vocal sounds. Can be made.

[0007] Further, in the sound image localization apparatus of the present invention, the mixing means uses the center localization signal component extracted from the input stereo signal, the sound image localization processing signal power output by the sound image localization processing means, and the center localization. You may have the structure added with the signal component except a signal zone | band.

With this configuration, the sound image localization apparatus of the present invention is such that the center localization signal component extracted from the input stereo signal force is the same as the sound localization processing signal force output by the sound image localization processing means. Since the center localization signal component that is localized in the center when output from a normal stereo speaker, which is often found in speech and vocal sounds, can be replaced with the sound image localization processing signal and output. This allows you to reproduce the natural surround sound without compromising the localization of the input signal component.

[0009] The sound image localization apparatus of the present invention further comprises center localization signal enhancement degree designation means for designating the enhancement degree of the center localization signal component, and the center localization signal designated by the center localization signal enhancement degree designation means. Based on the enhancement level of the signal component, the center The localization signal component detection means may have a configuration for detecting the center localization signal component.

With this configuration, the sound image localization apparatus of the present invention can reproduce the natural surround feeling by adjusting the enhancement degree of the center localization signal component according to the listener's preference.

[0011] Further, in the sound image localization apparatus of the present invention, the center localization signal component detection unit includes a correlation calculation unit that calculates a correlation between left and right signal components of the input stereo acoustic signal, and the correlation degree The center localization signal component may be detected based on the degree of correlation calculated by the calculation unit.

[0012] With this configuration, the sound image localization apparatus of the present invention can detect the center localization signal component, so that the localization feeling of the center localization signal component that is localized in the center direction when output from a normal stereo speaker is impaired. The natural surround sound can be reproduced.

[0013] Further, the sound image localization apparatus of the present invention includes correlation degree reference value setting means for setting a reference value of correlation degree, and the center localization signal component detection means is set by the correlation degree reference value setting means. The center localization signal component may be detected based on a reference value.

[0015] Further, the sound image localization apparatus of the present invention further includes center localization signal enhancement degree designation means for designating the enhancement degree of the center localization signal component, and the mixing means is detected by the center localization signal component detection means. A center localization signal extraction unit that extracts the center localization signal component from the input stereo signal based on the center localization signal band that has been extracted, and a gain adjustment unit that adjusts the gain of the extracted center localization signal component. The gain adjusting unit may be configured to adjust the gain of the center localization signal component based on the enhancement degree of the center localization signal component designated by the center localization signal enhancement degree designation means. ,.

With this configuration, the sound image localization apparatus of the present invention can reproduce the natural surround feeling by adjusting the enhancement degree of the center localization signal component according to the listener's preference. [0017] Further, in the sound image localization apparatus of the present invention, the center localization signal detection means is configured to reduce the center localization signal component band having substantially the same phase on the left and right sides from the input stereo acoustic signal by focusing on a middle to low range. You may have the structure detected as a zone | band.

[0018] With this configuration, the sound image localization apparatus of the present invention can detect the center localization signal component efficiently because the vocal signal band is often in the mid-low range.

In the sound image localization apparatus of the present invention, the mixing means includes a low-pass filter unit that extracts the input stereo acoustic signal power, the center localization signal band, and a frequency component lower than the center localization signal band. You may have.

[0020] With this configuration, the sound image localization apparatus of the present invention can extract a frequency component lower than the center localization signal band and the center localization signal band from the input stereophonic sound signal, and thus the localization feeling of the center localization signal component You can reproduce the natural surround sound without compromising the sound.

In the sound image localization apparatus of the present invention, the mixing unit includes a high-pass filter unit that extracts a frequency component higher than the sound signal localization processing signal force output from the sound image localization processing unit force and the center localization signal band. You may have.

[0022] With this configuration, the sound image localization apparatus of the present invention can extract a frequency component higher than the center localization signal band from the sound image localization processing signal, so that the natural localization feeling of the center localization signal component is not impaired. The surround feeling can be reproduced.

The invention's effect

[0023] According to the sound image localization apparatus of the present invention, there is provided a sound image localization apparatus capable of reproducing a natural surround feeling without impairing the localization feeling of sound components localized in the center direction, such as speech and vocal sounds. be able to.

Brief Description of Drawings

[0024] The features and advantages of the receiving terminal according to the present invention will become apparent from the following description in conjunction with the following drawings.

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

FIG. 2 is a block diagram of a sound image localization apparatus in a second embodiment of the present invention.

Explanation of symbols [0025] 1 L channel sound image localization processing means

2 R channel sound localization processing means

3, 5 High pass filter

4, 6 Low-pass filter

7, 8 delay circuit

9, 10 adder

11 Mixing processing unit (mixing means)

12 Correlation calculation means (correlation degree calculation unit)

13 Vocal band detection reference setting means (Correlation degree reference value setting means)

14 Vocal band detector (Center localization signal detector)

15 Vocal enhancement designation means (Center localization signal enhancement degree designation means)

21 L channel sound image localization processing means

22 R channel sound image localization processing means

23, 24 Calorie calculator

25 Delay circuit

26 Mixing processing unit (mixing means)

27 Correlation calculation means (correlation degree calculation part)

28 In-phase signal extraction means

29, 30 multiplier

31 Center localization signal extractor

32 Center localization signal enhancement level designation method

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)

A sound image localization apparatus according to a first embodiment of the present invention is shown in FIG.

In this embodiment, it is assumed that center localization signal components having substantially the same phase on the left and right that are localized in the center when output from a normal stereo speaker, which is common in speech and vocal sounds, are processed as “vocal signals”. [0028] As shown in FIG. 1, the sound image localization apparatus of the present embodiment is an L channel sound image localization processing means that localizes a sound image with respect to an input acoustic signal of the L (left side) channel from an input stereo acoustic signal. 1 and the sound image generated by the R channel sound image localization processing means 2 that localizes the sound image with respect to the input sound signal of the R (right) channel, the L channel sound image localization processing means 1 and the R channel sound image localization processing means 2 From the localization processing signal, a frequency component higher than the vocal signal band (hereinafter referred to as “vocal signal band”) is extracted, and the original input stereo acoustic signal power vocal signal band and a frequency component lower than the vocal signal band are extracted, The extracted vocal signal band and the original input signal with a frequency component lower than the vocal signal band are delayed by a predetermined time to obtain the vocal signal band. After mixing the phase with the V and sound image localization processing signals, the mixing processing unit 11 adds the respective signals, and the vocal signal band that detects substantially the same phase band on the left and right from the input stereo sound signal. A band detection unit 14 and vocal enhancement level designation means 15 for designating the enhancement level of the vocal signal detected by the vocal band detection unit 14 are provided. In this embodiment, the L-channel sound image localization processing means 1 and the R-channel sound image localization processing means 2 are used to localize a sound image at a predetermined opening angle wider than that of an actual speaker with respect to the input stereo sound signal. And In addition, the mixing processing unit 11 performs processing by determining a division frequency that is a frequency to be divided based on the vocal signal band detected by the vocal band detection unit 14.

[0029] The mixing processing unit 11 extracts a frequency component higher than the upper frequency of the vocal signal band detected by the vocal band detection unit 14 from the L channel sound image localization processing signal output from the L channel sound image localization processing means 1. High-pass filter (HPF) 3 and the R channel sound image localization processing signal output from the R channel sound image localization processing means 2 and the frequency higher than the upper frequency of the vocal signal band detected by the vocal band detector 14. A high-pass filter (HPF) 5 that extracts wave number components, a low-pass filter (LPF) 4 that extracts frequency components lower than the upper limit of the vocal signal band detected by the vocal band detector 14 from the L-channel input acoustic signal, and , Lower than the upper frequency limit of the vocal signal band detected by the vocal band detector 14 from the R channel input acoustic signal. A low pass filter (LPF) 6 for extracting a frequency component, output from the low pass filter 4 Delay circuit 7 that corrects the phase shift between the L channel input sound signal and the L channel sound localization signal output from the high pass filter 3 by delaying the L channel input sound signal, and the low pass filter 6 A delay circuit 8 that delays an R channel input acoustic signal output from the R channel and corrects a phase shift between the R channel input acoustic signal and the R channel sound image localization processing signal output from the high-pass filter 5; High pass filter 3 Adds the output of 3 outputs and the output of delay circuit 7 and outputs as an L-channel acoustic signal, and adds the output of no-pass filter 5 and the output of delay circuit 8 And an adder 10 for outputting as an R channel acoustic signal.

[0030] The vocal signal is often a highly correlated signal between the L channel and R channel input sound signals. For this reason, the vocal band detector 14 of the present embodiment detects the vocal signal band by detecting the correlation between the L channel acoustic signal and the R channel acoustic signal. That is, as shown in FIG. 1, the vocal band detector 14 converts the L-channel input sound signal and the R-channel input sound signal to the frequency axis by Fourier transform or the like, and calculates the correlation using the spectral similarity. Correlation calculation means 12 for calculating the correlation coefficient between the L channel input acoustic signal and the R channel input acoustic signal for each band, and the vocal signal band for the correlation coefficient for each band calculated by the correlation calculation means 12 Vocal band detection reference setting means 13 for setting a detection level as a reference to be determined. The vocal band detection unit 14 detects a band having a correlation coefficient equal to or greater than the reference value set by the vocal band detection reference setting unit 13 as a vocal signal band for each band calculated by the correlation calculation unit 12. It has become. At this time, since the vocal signal band is often in the mid-low range, detection may be performed by narrowing down the vocal signal band to be detected in advance to the mid-low range using a low-pass filter or the like.

The reference value set by the vocal band detection reference setting unit 13 is determined by the enhancement degree of the vocal signal designated by the vocal enhancement degree designation means 15. For this reason, the vocal signal enhancement degree designation means 15 can be used to designate the intensity of the vocal signal according to the listener's preference, for example. The vocal band detection reference setting means 13 sets a detection level for determining a vocal signal band according to the enhancement degree of the vocal signal designated by the vocal enhancement degree designation means 15. For example, the vocal emphasis preferred by listeners is high In this case, the vocal enhancement level specifying means 15 is operated to specify a high enhancement level of the vocal signal, the reference value of the correlation coefficient set by the vocal band detection reference setting means 13 is changed to a low value, and the vocal band detection unit 14 Increases the upper frequency division frequency that is considered the vocal signal band. On the other hand, if the listener's preferred vocal enhancement is low, the vocal enhancement specification means 15 is operated to specify a low vocal signal enhancement degree, and the correlation coefficient set by the vocal band detection reference setting means 13 is set. The reference value is changed to a higher value, and the upper frequency division frequency that the vocal band detection unit 14 determines as the vocal signal band is lowered. In this way, when the vocal emphasis level preferred by the listener is high, the vocal signal band is widened by operating the vocal emphasis specifying means 15 and lowering the threshold value for determining that the correlation is high. If the listener's preferred vocal emphasis level is low, adjust the emphasis level of the vocal signal by operating the vocal emphasis level specifying means 15 to increase the threshold value for determining that the correlation is high and narrowing the vocal signal band. can do.

[0032] As described above, according to the sound image localization apparatus of the present embodiment, the natural localization of the center localization signal that is localized in the center when output from a normal stereo speaker, which is common in speech and vocal sounds, is natural and is not impaired. The surround feeling can be reproduced.

[0033] In the present embodiment, the correlation calculating means 12 converts the L channel input acoustic signal and the R channel input acoustic signal into the frequency axis by Fourier transform or the like, and correlates with the similarity of the spectra. The correlation coefficient between the L channel input sound signal and the R channel input sound signal for each band is calculated, but the present invention is not limited to this. Any means may be used as long as the degree of correlation of the R channel input acoustic signal can be calculated. For example, the correlation calculating means 12 converts the L channel input acoustic signal and the R channel input acoustic signal into the frequency axis by Fourier transform or the like, calculates the correlation with the degree of spectral similarity, and then, for example, a predetermined 1Z3 octave band or the like. The correlation coefficient between the L channel input sound signal and the R channel input sound signal may be calculated for each frequency interval. In this case, it becomes possible to control the fine localization signal.

Furthermore, in the present embodiment, for ease of explanation, the vocal band detection unit 14 calculates the correlation between the left and right acoustic input signals in order to detect the vocal signal. Although the exit means 12 is used, the present invention is not limited to this. It goes without saying that the vocal band detector 14 may use any means that can detect vocal signal components having substantially the same phase on the left and right.

In the present embodiment, the center localization signal is extracted from the original L channel input acoustic signal and the R channel input acoustic signal, and the L channel and R channel sound image localization processing signals are center localization. Although it has been described that the calculation is performed on the signal component from which the frequency component of the signal band is removed, the present invention is not limited to this. For example, if the center localization signal is output from a stereo speaker, the sound component tends to be reduced by the sound image localization process. Therefore, the gain of the center localization signal extracted from the L channel input acoustic signal and the R channel input acoustic signal is extracted. May be adjusted and added to the L channel and R channel sound image localization processing signals. This will be described in the next embodiment.

(Second embodiment)

A sound image localization apparatus according to the second embodiment of the present invention is shown in FIG.

In the present embodiment, it is assumed that an acoustic signal component having substantially the same phase on the left and right, which is localized in the center when output from a normal stereo speaker, which is common in speech and vocal sounds, is processed as a “center localization signal”.

[0037] As shown in FIG. 2, the sound image localization apparatus of the present embodiment is an L channel sound image localization processing means for localizing a sound image with respect to an input sound signal of the L (left) channel from an input stereo sound signal. 21 and R channel sound image localization processing means 22 that localizes the sound image relative to the input sound signal of the R (right) channel, and a center localization signal that is substantially in phase with each of the input sound signals of the L channel and the R channel. Center localization signal extraction unit 31 and mixing for adding the center localization signal extracted by center localization signal extraction unit 31 to the signals subjected to sound image localization processing by L channel sound image localization processing means 21 and R channel sound image localization processing means 22 A processing unit 26 and a center localization signal enhancement level specifying means 32 for specifying the enhancement level of the center localization signal detected by the center localization signal extraction unit 31 are provided. is doing. In this embodiment, it is assumed that the L channel sound image localization processing means 21 and the R channel sound image localization processing means 22 localize a sound image at a predetermined opening angle wider than the actual speaker with respect to the input stereo sound signal. [0038] Here, the center localization signal extraction unit 31 converts the L-channel input acoustic signal and the R-channel input acoustic signal into the frequency axis by Fourier transform or the like, calculates the correlation based on the similarity of the spectrum, and Correlation calculating means 27 for calculating the correlation coefficient between the L channel input acoustic signal and the R channel input acoustic signal, and when the number of correlations calculated by the correlation calculating means 27 is equal to or greater than a reference threshold, The in-phase signal extraction means 28 for extracting the original L-channel input sound signal and the R-channel input sound signal of each of the R-channel sound signals, and the L-channel sound signal extracted by the in-phase signal extraction means 28 A multiplier 29 that adjusts the gain of the L-channel acoustic signal by multiplying by a predetermined multiplication value, and the R channel extracted by the in-phase signal extraction means 28. And a multiplier 30 that adjusts the gain of the R channel sound signal by multiplying the sound signal by a predetermined multiplication value.

[0039] Further, the mixing processing unit 26 delays the center localization signal extracted by the center localization signal extraction unit 31, and the center localization signal and the L channel subjected to localization processing by the L channel sound image localization processing means 21 A delay circuit 25 for correcting the phase shift between the sound image localization processing signal and the R channel sound image localization processing signal localized by the R channel sound image localization processing means 22, and the center localization corrected in this way by the delay. The signal is composed of an adder 23 for adding the L channel sound image localization processing signal to the signal and an adder 24 for adding the R channel sound image localization processing signal.

The multiplication value multiplied by the multipliers 29 and 30 is determined by the enhancement degree of the center localization signal designated by the center localization signal enhancement degree designation means 32. For example, the center localization signal enhancement level designation means 32 can designate the intensity of the center localization signal according to the listener's preference. For example, when the vocal enhancement level preferred by the listener is high, the center localization signal enhancement level designation is performed. The means 32 is operated to specify a high degree of enhancement of the center localization signal, and the multiplication value multiplied by the multipliers 29 and 30 is increased. In this way, the gain of the center localization signal extracted by the center localization signal extraction unit 31 is increased, and the level of the center localization signal input to the adders 23 and 24 is increased. Conversely, if the listener's preferred vocal enhancement level is low! ヽ, operate the center localization signal enhancement level designation means 32 to specify a low level of center localization signal enhancement level, and multipliers 29 and 30 multiply Reduce the multiplication value. In this way, the gain of the center localization signal extracted by the center localization signal extraction unit 31 is reduced, and the level of the center localization signal input to the adders 23 and 24 is lowered.

[0041] As described above, according to the sound image localization apparatus of the present embodiment, the natural localization of the center localization signal that is localized to the center when output from a normal stereo speaker, which is common in speech and vocal sounds, is natural and is not impaired. The surround feeling can be reproduced.

[0042] In the present embodiment, the correlation calculating means 27 converts the L channel input acoustic signal and the R channel input acoustic signal into the frequency axis by Fourier transform or the like, and correlates with the similarity of the spectra. The correlation coefficient between the L channel input sound signal and the R channel input sound signal for each band is calculated, but the present invention is not limited to this. Any means may be used as long as the degree of correlation of the R channel input acoustic signal can be calculated. For example, the correlation calculating means 27 converts the L-channel input acoustic signal and the R-channel input acoustic signal into the frequency axis by Fourier transform or the like, calculates the correlation based on the similarity of the spectrum, and then, for example, a predetermined 1Z3 octave band or the like. The in-phase signal extraction means 28 calculates the correlation coefficient between the L channel input acoustic signal and the R channel input acoustic signal for each frequency section, and the correlation coefficient thus calculated by the correlation calculation means 27 is equal to or greater than the reference threshold value. For each frequency section, the original L channel input sound signal and R channel input sound signal of the L channel sound signal and the R channel sound signal force may be extracted. In this case, finer control of the center localization signal is possible.

[0043] Further, in the above-described embodiment, for ease of explanation, the center localization signal extraction unit 31 calculates the correlation between the left and right acoustic input signals in order to detect and extract the center localization signal. Although the means 27 is used, the present invention is not limited to this. The center localization signal extraction unit 31 may use any means as long as it can detect signal components having substantially the same phase on the left and right.

Industrial applicability

[0044] The sound image localization device that is useful in the present invention is a natural smooth sound that does not impair the sense of localization of the sound component that is localized at the center when output with normal stereo sound, which is often found in speech and vocal sounds. Reproducing the sense of sound has a special feature and is useful as a sound image localization device that localizes the sound image of a sound output from a stereo speaker such as a mobile phone or a portable speaker.

Claims

The scope of the claims

[1] sound image localization processing means for performing sound image localization processing on an input stereo sound signal and outputting a sound image localization processing signal;

Center localization signal detecting means for detecting a center localization signal component band having substantially the same phase on the left and right as the center localization signal band from the input stereo sound signal;

Based on the center localization signal band detected by the center localization signal component detection means, the center localization signal component is extracted from the input stereo signal based on the center localization signal band, and the sound image localization processing output by the sound localization processing means is output. A sound image localization apparatus comprising mixing means for adding to a signal.

[2] The mixing unit adds the center localization signal component extracted from the input stereo signal to a signal component obtained by removing the center localization signal band from the sound image localization processing signal output by the sound image localization processing unit. The sound image localization apparatus according to claim 1, wherein

[3] Furthermore, a center localization signal enhancement degree designation means for designating the enhancement degree of the center localization signal component is provided,

2. The center localization signal component detection unit detects the center localization signal component based on the enhancement level of the center localization signal component designated by the center localization signal enhancement degree designation unit. The sound image localization apparatus described.

[4] The center localization signal component detection means includes a correlation degree calculation unit that calculates a correlation degree between right and left signal components of the input stereo sound signal, and the correlation degree calculated by the correlation degree calculation unit is calculated. The sound image localization apparatus according to claim 1, wherein the center localization signal component is detected based on the center localization signal component.

[5] Correlation degree reference value setting means for setting a correlation degree reference value is provided, and the center localization signal component detection means is based on the reference value set by the correlation degree reference value setting means. 5. The sound image localization apparatus according to claim 4, wherein a signal component is detected.

[6] Furthermore, a center localization signal enhancement degree specifying means for specifying the enhancement degree of the center localization signal component is provided, The mixing means includes a center localization signal extraction section for extracting the center localization signal component based on the center localization signal band detected by the center localization signal component detection means and the input stereo signal force, and the extracted It has a gain adjustment unit that adjusts the gain of the center localization signal component,

The gain adjustment unit adjusts the gain of the center localization signal component based on the enhancement level of the center localization signal component specified by the center localization signal enhancement level designation unit. The sound image localization apparatus described.

[7] The center localization signal detection means is characterized by detecting a center localization signal component band having substantially the same phase on the left and right as a center localization signal band from the input stereophonic sound signal in the middle to low range. The sound image localization apparatus according to claim 1.

8. The mixing unit according to claim 1, wherein the mixing unit includes a low-pass filter unit that extracts a frequency component lower than the center localization signal band and the center localization signal band from the input stereo sound signal. Sound image localization device.

9. The mixing device according to claim 2, wherein the mixing unit includes a high-pass filter unit that extracts a frequency component whose sound image localization processing signal power output from the sound image localization processing unit is higher than the center localization signal band. Sound image localization device.