JPH099398A - Sound image localization device - Google Patents

Abstract

PURPOSE: To provide the sound image localization device in which more detailed adjustment than that of a conventional device is attained with less data amount and simpler hardware than the conventional device. CONSTITUTION: In the sound image localization device using filter sections 2-1, 2-2 so as to process a digital acoustic signal from a sound source 1 so as to localize an imaginary sound image, the filters 2-1, 2-2 are made up of arithmetic means 3-1, 3-2 conducting convolution arithmetic operation between the digital acoustic signal and a transfer function with a prescribed frequency characteristic, delay means 4-1, 4-2 delaying the digital acoustic signal, and equalizers 5-1, 5-2 applying filter processing to the digital acoustic signal, and a controller 20 controls the delay in the delay means 4-1, 4-2 and the frequency characteristic of the equalizers 5-1, 5-2 based on the angle information of the imaginary sound image and the head shape information of a listener 11.

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 device for realizing a virtual sound image in, for example, AV (audio / visual) equipment.

[0002]

2. Description of the Related Art In recent years, in the fields of video, audio and games, it has been desired to develop a device that realizes virtual reality. In the field of acoustics, techniques such as sound field control and sound image control have been particularly advanced, and devices that actually realize these techniques are being actively developed.

FIG. 9 is a block diagram of a conventional sound image localization apparatus. In FIG. 9, 1 is a sound source for outputting a digital sound signal, and 102-1 and 102-2 are given digital sound signals from the sound source 1. The left and right filter parts,
Each of the filter units 102-1 and 102-2 has an arithmetic unit 10
3-1 and 103-2 are provided. Reference numeral 13 is a sound image angle information input means for inputting the angle of the virtual sound image to the controller, and 120 is a calculation means 103-1 for reading the tap coefficient from the memory based on the angle information of the virtual sound image from the sound image angle information input means 13. , 103-2, and controllers 9-1, 9-2 are arithmetic means 103-1, 103-2.
The left and right digital audio signals from the
An amplifier that amplifies each after D / A conversion by the converter,
Reference numeral 10 is a headphone that outputs acoustic signals from the amplifiers 9-1 and 9-2 as sound, and 11 is a listener.

As shown in FIG. 10, the controller 120 reads the tap coefficient from the coefficient memory 121 based on the angle information from the sound image angle information input means 13 and calculates the filter units 102-1 and 102-2. Means 103
-1, 103-2 are provided to the calculation means 103.
Reference numerals -1, 103-2 are, for example, configured by FIR filters, and perform convolutional calculation between the acoustic signal and the tap coefficient.

In such a sound image localization apparatus, as shown in FIG. 9, the operation principle when the virtual sound image 14 is localized to the left front of the listener 11 will be described.

Here, h1 (t) is a response between the headphone 10 and the left ear of the listener 11 (accurately, a response at the position of the eardrum when an impulse is input to the headphone. Head-related transfer function (hereinafter, referred to as an impulse response in order to explain in the time domain. However, similar results can be obtained in the frequency domain), h2 (t), and And the impulse response at the position of the listener 11's right ear, S (t) is the acoustic signal from the sound source 1, h3 (t) is the impulse response at the virtual image and the position of the listener 11's left ear, h4 (t) Is the impulse response at the position of the virtual sound image and the right ear of the listener 11, hR
Let (t) be the impulse response of the filter section and hL (t) be the impulse response of the filter section.

When the signal S (t) from the sound source 1 is emitted from the headphones 10, the sound that reaches the ear of the listener 11 is the left ear L (t)

[0008]

[Equation 1]

In the right ear R (t),

[0010]

[Equation 2]

It is expressed as

Further, the impulse response and the signal S (t) are considered as digital signals whose time is discrete, and L (t) → L (n) R (t) → R (n) h1 (t) → h1 ( n) h2 (t) → h2 (n) hR (t) → hR (n) hL (t) → hL (n) S (t) → S (n) (n is actually nT, T is the sampling time In general, T is abbreviated, and n is a natural number. Then, (Equation 1) and (Equation 2) are as follows.

[0013]

(Equation 3)

[0014]

[Equation 4]

When radiated from the real speaker located at the position of the virtual sound image 14, the sound reaching the ear of the listener 11 is the left ear L '.
(t)

[0016]

(Equation 5)

In the right ear R '(t)

[0018]

(Equation 6)

Is expressed as

If the head-related transfer functions are equal, it is assumed that the sound is heard from the same direction (this assumption is generally correct).

[0021]

(Equation 7)

At the time,

[0023]

(Equation 8)

[0024]

[Equation 9]

At the time,

[0026]

(Equation 10)

Therefore, in order to allow the listener 11 to hear from the position of the virtual sound image 14 using the headphones 10, in order to satisfy (Equation 8) and (Equation 10), h
It suffices to determine l (n) and hR (n).

For example, if (Equation 8) and (Equation 10) are rewritten in the frequency domain expression, the convolution operation is replaced by multiplication, and thereafter, the respective impulse responses are FFT'ed into a transfer function. Since the transfer functions other than the two transfer functions of the FIR filter are obtained by measurement, the transfer function of the FIR filter can be obtained from these two expressions.

HL (n), hR thus determined
(n) and the signals S (n) and hL from the headphones 10 are used.
(n) and the signals S (n) and hR (n) from the headphones 10.
By radiating the convoluted and
Makes it possible to make the user feel that a sound is coming from the position of the virtual sound image 14.

In this way, the sound image can be localized at any position.

[0031]

However, in the configuration of the conventional example, since the tap coefficient data of the filter section is required for each angle at which the sound image is localized, it is necessary to realize the sound images at all angles. , Very large amount of data is required, and very large amount of memory is required to store this data.

Further, since the data of the filter section is switched for each angle, the processing becomes complicated. Furthermore, since the size of the head and the like vary from person to person, it is necessary to prepare many patterns of data for the filter section in order to match the listener.

The present invention has been made in view of the above technical problems, and it is possible to reduce the amount of data and the amount of calculation in the filter section and to realize a localization feeling and sound quality comparable to those of the conventional example. The purpose is to

[0034]

In order to achieve the above-mentioned object, the present invention is configured as follows.

That is, the sound image localization apparatus of the present invention comprises a calculating means for performing a convolution calculation of a digital acoustic signal, a delay means for delaying the digital acoustic signal, a filter means for filtering the digital acoustic signal, and an angle of the virtual sound image. And a control means for controlling the delay amount of the delay means and the frequency characteristic of the filter processing by the filter means based on the information.

In the present invention, the calculation means performs a convolution calculation of the digital acoustic signal and the transfer function having a predetermined frequency characteristic, and this transfer function transfers between the listener and a predetermined virtual sound image. It substantially matches the function.

In the present invention, the filter means is an equalizer whose frequency characteristic is controlled by the control means, and the frequency characteristic has a predetermined cutoff frequency.

Further, in the present invention, the control means may control the delay amount and the frequency characteristic based on the angle information and the personal information of the listener, and as the personal information of the listener, You may use the size information of a listener's head.

[0039]

According to the sound image localization apparatus of the present invention, the coefficient of the convolution calculation is not controlled based on the angle information of the virtual sound image as in the prior art, but based on the angle information of the virtual sound image,
Since the delay amount by the delay unit is controlled to control the interaural time difference and the frequency characteristic of the filter unit to adjust the interaural level difference to localize the virtual sound image, the data amount is smaller than the conventional one, and A sound image localization device can be realized with relatively simple hardware.

Further, by controlling the delay amount and the frequency characteristic based on the personal information such as the size information of the listener's head in addition to the angle information of the virtual sound image, the data amount is smaller than the conventional one. Moreover, the localization of the virtual sound image can be realized according to the listener with relatively simple hardware.

[0041]

Embodiments of the present invention will now be described in detail with reference to the drawings.

(Embodiment 1) FIG. 1 is a block diagram of a sound image localization apparatus according to an embodiment of the present invention, in which parts corresponding to the above-mentioned conventional example are designated by the same reference numerals.

The sound image localization apparatus of this embodiment has a sound source 1 for outputting a digital sound signal and left and right filter sections 2-1 and 2-2 to which the digital sound signal from the sound source 1 is applied.
-2, and an amplifier 9-which amplifies the digital acoustic signal, which has been subjected to signal processing by the filter units 2-1 and 2-2 as described later, after D / A conversion by a D / A converter (not shown).
1, 9-2, headphones 10 for outputting the acoustic signals of the amplifiers 9-1, 9-2 as sounds, and sound image angle information input means for inputting angle information corresponding to the position at which the virtual sound image is localized to the controller 20. 13 and personal information input means 1 for inputting personal information corresponding to the size of the head of the listener 11 based on the opening degree of the headphones 10 to the controller 20.
2 and a controller 20 as a control unit that controls the filter units 2-1 and 2-2 based on the angle information and the personal information.

The filter units 2-1 and 2-2 for processing the digital audio signal include arithmetic means 3-1 and 3-2 which are FIR filters and delay means 4-1 and 4 which delay the digital audio signal. -2, and an equalizer 5-1 as a filter means for filtering the digital audio signal.
5-2 and.

In the following description, one of the filter units 2-1 and 2-2 2-1 will be described, but the other one is also the same.

The calculating means 3-1 performs a convolution calculation of the digital acoustic signal and the transfer function having a predetermined frequency characteristic, and the predetermined frequency characteristic is a predetermined real speaker as a predetermined virtual sound image. , It is set so as to have a head-related transfer characteristic between predetermined listeners. This predetermined real speaker is at an arbitrary angle diagonally forward of the listener 11 (excluding front front and front rear), and the predetermined listener is
Optional. However, the certain angle of the predetermined actual speaker and the head shape information of the predetermined listener used here are the reference data of the controller 20, and are the reference angle and the reference information, respectively.

As shown in FIG. 2, the calculating means 3-1 is composed of an FIR filter, and this FIR filter is provided with delay devices 61 to 64, multipliers 65 to 68 and an adder 69. The frequency characteristics are as described above.
It is set so as to have a head-related transfer characteristic between a predetermined real speaker and a predetermined listener.

The actual listener 1 from the personal information input means 12
The controller 20 to which the head shape information of No. 1 and the angle information of the virtual sound image 14 are input from the sound image angle information input unit 13 is input to the coefficient memory 21 as shown in FIG. Therefore, the coefficient is set in the delay means 4-1 and the equalizer 5-1 to control the delay amount of the delay means 4-1 and the frequency characteristic of the equalizer 5-1. The head shape information of the listener 11 is a parameter indicating the size of the listener's head, and the angle information is a parameter indicating the angle at which the virtual sound image viewed from the listener 11 should be localized.

In this embodiment, the delay means 4-1 delays by controlling writing / reading of the digital acoustic signal to / from the memory.

As shown in FIG. 4, the equalizer 5-1 of this embodiment is a biquad IIR filter having a two-stage configuration. The IIR filter includes delay devices 31 to 31.
4, multipliers 35 to 39, and an adder 40.

Next, the control of the controller 20 when the virtual sound image is localized in the range from 0 degree to 90 degrees clockwise with 0 degree in front of the listener 11 will be described with reference to FIGS. 5 and 6. Explain. It should be noted that the description will be made for 0 degrees to 90 degrees, but 0 degrees to −90 degrees and 90 degrees to 1 degree.
The same processing can be performed for 80 degrees, -90 degrees to -180 degrees.

First, the angle information of the virtual sound image and the listener 11
A method for determining the inter-aural time difference of the listener 11 from the head shape information will be described with reference to FIG. In this embodiment, of the head shape information, mainly the distance information between both ears, that is, the size of the head is used, but the present invention is not limited to this, and other information such as listening information is used. The same effect can be obtained by using the sex and age of the person 11.

In FIG. 5, the horizontal axis represents the angle of the virtual sound image,
The vertical axis represents the binaural time difference to be set, and the solid line a1
Indicates the characteristics of the predetermined listener described above, and the broken line b1 indicates the characteristics of an example of the actual listener 11.

The distance between the ears of the listener changes depending on the size of the head, so that the transmission characteristics from the virtual sound image to the listener have different time differences between the ears, and the head information of the listener 11 is obtained. As a result, the characteristics shown in FIG. 5 can be obtained.

For example, the difference in the interaural time difference between the predetermined listener and the actual listener 11 at the angle x of the virtual sound image is d1.
It is. In this embodiment, the controller 20 controls the delay amount of the delay unit 4-1 so that the listener 11 can recognize the virtual sound image so that the difference d1 is eliminated, that is, a1 of the predetermined listener. The time difference of the characteristics is adjusted to be the time difference of the characteristics of b1 of the actual listener.

As described above, the controller 20
The interaural time difference set in the delay unit 4-1 is uniquely determined from the head information from the personal information input unit 12 and the angle information of the virtual sound image.

In this embodiment, the relationship between the angle of the virtual sound image and the binaural time difference is shown as a straight line.
The relationship between the angle of the virtual sound image and the binaural time difference can be processed in the same manner even in the case of an arbitrary function having the size of the head as a parameter.

Next, a method of determining the interaural level difference of the listener 11 from the angle information of the virtual sound image will be described with reference to FIG.

FIG. 6 is a diagram showing a change in the ratio of the interaural level difference with respect to the change in the angle of the virtual sound image with reference to the angle of the predetermined predetermined real speaker, that is, the characteristic of the reference angle. Yes, the horizontal axis represents the frequency and the vertical axis represents the level difference to be set.

In FIG. 6, a2 shows the characteristic of the reference angle, b2 shows the ratio of the interaural level difference at the angle x1, and c2 shows the ratio of the interaural level difference at the angle x2, for example. There is.

In the region below the cutoff frequency f, the level difference is not recognized. Therefore, in this embodiment, in order to simplify the circuit configuration, the characteristic b2 below the characteristic a2 is set to a predetermined characteristic. The low-pass, low-frequency emphasis characteristic b21 having the cutoff frequency f, and the characteristic c2 above a2 are approximate to the high-pass, high-frequency emphasis characteristic c21. Here, the predetermined cutoff frequency f is 500 Hz or more 2
It is an arbitrary frequency below kHz. It should be noted that d2 represents the difference in level difference in the region above the cutoff frequency.

In this way, the frequency characteristic of the equalizer 5-1 is controlled according to the angle information of the virtual sound image.

A filter having such characteristics can be realized by a low-order IIR filter.

That is, in the prior art, a large amount of angle adjustment data had to be prepared for each angle, but in the present invention, similar processing is performed with a small amount of data and a large amount of calculation by using a low-order IIR filter. realizable.

As the angles x1 and x2 change, the characteristic of the interaural level difference changes from a2 to b2 or c2. Conventionally, it was necessary to prepare data for each angle, and complicated processing such as switching of these data was required, but in the present invention, only the magnitude of the level difference difference d2 is adjusted according to the angle. Therefore, it can be realized with less data, calculation amount and hardware,
Even finer angle adjustment can be easily performed.

As described above, the filter units 2-1 and 2-2
-2, the output processed through amplifiers 9-1, 9-2,
By reproducing with the headphones 10, the sound image can be localized to the listener 11 at an arbitrary angle.

In this way, the sound image data corresponding to the predetermined angle of the virtual sound image of the predetermined listener is used, and from the head information from the personal information input means 12 and the angle information of the virtual sound image, the appropriate interaural distance is obtained. By setting the time difference and the interaural level difference and adjusting the delay amount and the level difference, it is possible to realize a sound image localization apparatus capable of finer adjustment than before with a smaller amount of data and simple hardware than ever before.

As a result, conventionally, a large amount of angle adjustment data had to be prepared for each angle, but in the present invention, similar processing can be realized with a small amount of data and a large amount of calculation.

When the listener changes, conventionally, it was necessary to prepare the data of each listener or use the representative value. Therefore, the circuit scale becomes large,
If the value does not match the representative value, the angle may shift. In the present invention, since the listener can be finely adjusted according to the size of the listener's head, it is possible to easily cope with changes in the listener with a small amount of data.

In this embodiment, the head information of the listener 11 is input to the personal information input means 12 based on the degree of opening of the headphones 10. However, the present invention is not limited to this, and the personal information is externally input. You may make it possible to input to the input means 12.

In this embodiment, the digital acoustic signal is processed in the order of the arithmetic means 3-1, the delay means 4-1, and the equalizer 5-1. However, the present invention is not limited to this, and the arbitrary order is used. May be processed in.

In this embodiment, two delay means for adjusting the interaural time difference and two equalizers for adjusting the interaural level difference are provided corresponding to the left and right ears, but as another embodiment of the present invention. Alternatively, it may be provided in only one of them to adjust the interaural time difference and the interaural level difference.

(Embodiment 2) In the above-mentioned embodiment, the interaural level difference of the listener 11 is determined only from the angle information of the virtual sound image, but as another embodiment of the present invention, the angle information of the virtual sound image and The binaural level difference may be determined based on both the head shape information of the listener 11. That is, FIG. 7 is a diagram showing a change in the ratio of the interaural level difference with respect to the change in the size of the head based on the characteristics of the head shape information (reference information) of the predetermined listener described above. The horizontal axis represents frequency and the vertical axis represents level difference to be set.

Further, a3 represents a characteristic in the head shape information of the predetermined listener, and b3 represents an interaural level difference in the head shape information of an example of a listener whose head is larger than that of the predetermined listener. C3 represents the ratio of the interaural level difference in the head shape information of an example of a listener whose head is smaller than a predetermined listener, and d3 represents the difference in the level difference in the region above the cutoff frequency. ing.

In the present invention, in order to simplify the circuit configuration, in FIG. 7, the characteristic b3, which is a characteristic lower than a3 with reference to the characteristic a3, is a low-pass and low-frequency emphasis characteristic having a predetermined cutoff frequency f. b31 has a characteristic c above a3
3 is close to the high-pass, high-frequency emphasis characteristic c31. Here, the predetermined cutoff frequency f is an arbitrary frequency of 500 Hz or more and 2 kHz or less.

The characteristic for determining the interaural level difference of the listener 11 from the angle information of the virtual sound image is as shown in FIG. 6 as described above, and the interaural level difference is determined from the head shape information of the listener 11. Since the characteristic for determining is that shown in FIG. 7, it may be controlled so as to be a combined characteristic of the two characteristics shown in FIGS. 6 and 7, as shown in FIG. Note that FIG. 8 shows a characteristic obtained by combining the characteristic b2 of FIG. 6 and the characteristic b3 of FIG.

A filter having such characteristics can be realized by a low-order IIR filter as in the above embodiment.

As a result, only the magnitude of the level difference d4 as shown in FIG. 8 needs to be adjusted in accordance with the angle information of the virtual sound image and the personal information of the listener, and a smaller amount of data than the conventional example is used. It will be possible to realize with the quantity and the hardware.

The other structure is the same as that of the above-mentioned embodiment.

[0080]

As described above, according to the present invention, the delay amount and the frequency characteristic of the digital sound signal corresponding to the predetermined listener and the predetermined virtual sound image angle are determined based on the virtual sound image angle information. By controlling the interaural time difference and the interaural level difference to localize the virtual sound image, the sound image localization device is realized with a relatively simple hardware with a smaller amount of data than before. it can.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a configuration of a computing means of the embodiment of FIG.

FIG. 3 is a block diagram of a main part of the embodiment of FIG.

FIG. 4 is a diagram showing a configuration of an equalizer of the embodiment shown in FIG.

5 is a characteristic diagram of the interaural time difference of the embodiment of FIG.

6 is a characteristic diagram of the binaural level difference of the embodiment of FIG.

FIG. 7 is a characteristic diagram of interaural level difference according to another embodiment of the present invention.

8 is a characteristic diagram of the binaural level difference of the embodiment of FIG.

FIG. 9 is a block diagram of a conventional example.

10 is a block diagram of a main part of the conventional example of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 sound source 2-1 and 2-2 filter part 3-1, 3-2 computing means 4-1 and 4-2 delay means 5-1 and 5-2 equalizer (filter means) 10 headphones 11 listener 12 personal information input Means 13 Sound image angle information input means 20 Controller (control means)

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Shoji Matsumoto 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (72) Akihisa Kawamura, 1006 Kadoma, Kadoma City, Osaka Matsushita Electric Industrial Co., Ltd.

Claims (9)

[Claims] 1. A calculation means for performing a convolution calculation of a digital sound signal, a delay means for delaying the digital sound signal, a filter means for filtering the digital sound signal, and the delay means on the basis of angle information of a virtual sound image. And a control means for controlling the frequency characteristic of the filter processing by the filter means, and the sound image localization apparatus. 2. The delay means delays the digital acoustic signal convoluted by the computing means, and the filter means filters the digital acoustic signal delayed by the delay means. The sound image localization apparatus according to claim 1. 3. The sound image localization apparatus according to claim 1, wherein the control unit controls the delay amount and the frequency characteristic based on the angle information and the personal information of the listener. 4. The calculation means performs a convolution calculation of the digital acoustic signal and a transfer function having a predetermined frequency characteristic, and the transfer function is a transfer between a listener and a predetermined virtual sound image. The sound image localization apparatus according to claim 1, wherein the sound image localization apparatus substantially matches the function. 5. The sound image localization device according to claim 4, wherein the predetermined virtual sound image is a virtual sound image at a predetermined angle other than the front front and the front rear of the listener. 6. The predetermined angle is an arbitrary angle between 20 degrees and 80 degrees or 280 degrees and 340 degrees clockwise when the front of the listener is 0 degrees in front of the listener. The sound image localization apparatus according to claim 5. 7. The sound image localization apparatus according to claim 1, wherein the personal information of the listener is size information of the listener's head. 8. The sound image according to claim 1, wherein the filter means is an equalizer whose frequency characteristic is controlled by the control means, and the frequency characteristic has a predetermined cutoff frequency. Localization device. 9. The sound image localization apparatus according to claim 8, wherein the predetermined cutoff frequency is a frequency of 500 Hz or more and 2 kHz or less.