JPH08191225A - Sound field reproducing device - Google Patents

Abstract

PURPOSE: To attain natural senses of sound image positioning, distance and expansion in all directions by the use of a headphone or an innerphone. CONSTITUTION: A sound signal from a signal input means 1 is applied to an A/D converter 2, which converts the sound signal into a digital signal at a sampling frequency fs. At the time of inputting the digital signal, a signal processing means 3 executes operation for sound image positioning by the use of an FIR filter. In this case, a convolution coefficient is over-sampled and a tap position for multiplying the convolution coefficient is finely controlled in accordance with a distance difference between both ears of a listener. Thus the convolution coefficient is controlled in accordance with the shape and size of the head part of the listener and positional relation between the listener and a sound image. Thus the operated sound signal is applied to D/A converters 4a, 4b and D/A converted signals are outputted from speakers 6a, 6b of a headphone through amplifiers 5a, 5b.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an AV (audio / visual) device, in order to reproduce sound with a natural localization and presence for an arbitrary listener, in order to reduce the distance between individual ears and the head. The present invention relates to a sound field reproducing device that takes the shape of a part into consideration.

[0002]

2. Description of the Related Art In recent years, in the field of multimedia centering on video and audio, it has been desired to develop hardware that can reproduce three-dimensional stereophonic sound in association with the three-dimensional computer graphics of video. There is. In addition, interactive devices such as home-use game machines for TVs and personal computers are becoming popular. In order to deal with this, in the sound field reproduction using not only a two-channel stereo speaker but also an ear speaker, headphones, and an inner phone, development is underway to interactively control the localization, distance, and expanse of sound.

FIG. 11 is a block diagram showing an example of a conventional sound field reproducing apparatus using a speaker. In the figure,
The signal input means 101 is an input means for inputting an audio signal, and its output is given to the amplifiers 103 a and 103 b via the filter 102. Filter 102 is 2
It controls the frequency characteristics, volume, and time delay of the channel signal, and is controlled by the sound image position input means 104. The sound image position input means 104 is means for inputting the sound image position in order to control the listener's sense of localization of the sound. Amplifier 1
The audio signal amplified by 03a and 103b is the speaker 10
5a, 105b.

The operation of the conventional sound field reproducing apparatus having such a configuration will be described. The audio signal input by the signal input unit 101 controls the filter 102 to reproduce the audio from the speakers 105a and 105b with the frequency characteristic, volume, and time delay according to the position of the sound input input by the sound image position input unit 104. To do. When the left ear has a longer time delay than the right ear and the sound pressure level is low due to the frequency characteristics in the middle and high frequencies, it is generally felt that the sound image is on the right side. Filter 10
2 performs such control to control the listener's sense of sound image localization. The signal controlled by the filter 102 is the amplifier 1
Amplified by 03a, 103b, speaker 105a,
It is reproduced by 105b. By performing such signal processing, it is possible to control the sound localization.

[0005]

However, with the above configuration, it is difficult to control a subtle change in characteristics due to a difference in the shape of the listener's head and ear even when listening to the reproduced sound. It was In particular, in the case of headphones or an inner phone inserted into the ear canal, it was not possible to realize the forward out-of-head feeling, localization feeling and distance feeling in the sound image.

The present invention has been made in view of the above-mentioned problems of the related art, and a sound which gives a natural sound image localization feeling and distance feeling in all directions to any listener is obtained. It is intended to provide a field reproducing device.

[0007]

According to the present invention, an A / D converter for converting an input voice signal into a digital signal and a digital signal converted by the A / D converter are input, and the head shape of a listener or Signal processing means for performing signal processing so as to obtain an acoustic signal according to the size, the position and distance between the listener and the sound image,
Input means for inputting the shape and size of the listener's head, and the position of the listener and the sound image, and setting the parameters so that the acoustic characteristics according to the contents input by the input means are obtained,
A parameter control means for giving a parameter to the signal processing means, a D / A converter for converting the signal output from the signal processing means into an analog signal, an amplifier for amplifying the output of the D / A converter, and an output of the amplifier are provided. A sound field reproducing device comprising left and right acoustic transducers for performing acoustic conversion, wherein the signal processing means localizes a sound image by convolution calculation according to the positional relationship between the listener and the sound image input by the input means. The sound image localization means for inputting and the reference convolution coefficient as parameters from the parameter control means via the sound image localization means, these parameters are converted into time-series discretized signals, and the sampling frequency of the A / D converter or higher. Oversampling means for oversampling the parameter with the value of And a parameter output from the filter means are down-sampled with a value equal to or lower than the sampling frequency of the over-sampling means, and the obtained time-series signals are parallel-converted into respective convolution coefficients, and the value thereof is obtained. Down-sampling means for giving to the sound image localization means as a corrected convolution coefficient.

[0008]

According to the present invention having such features, the head shape and size of the listener and the position and distance between the listener and the sound image are input and set by the input means. When an audio signal is input via the signal input means, it is converted into a digital signal by the A / D converter and input to the signal processing means. The parameter control unit selects a parameter so that the acoustic characteristic according to the content set by the input unit can be obtained. Then, a parameter that matches the established condition is given to the signal processing means as a convolution coefficient. The signal calculated by the signal processing circuit is converted into an analog signal by the D / A converter, and the sound is output from the acoustic converter via the amplifier.

Particularly, in the signal processing means, the sound image is localized according to the head shape and size of the listener input by the input means and the positional relationship between the listener and the sound image, and the output thereof is oversampled to obtain a predetermined frequency characteristic. After filtering with
An audio signal is output by down-sampling at a value close to the original sampling frequency. At this time, in order to reflect the subtle binaural gap difference of an arbitrary listener in the output signal, oversampling is performed at a necessary sampling frequency. Further, when the oversampled signal is applied to the fill means, a component having a component of 1/2 or less of the sampling frequency of the A / D converter passes through. This reduces distortion when down-sampling by the down-sampling means. By processing in this way, a definite binaural gap difference can be dealt with for any listener, and natural sound image localization and sound field reproduction can be obtained in all directions.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A sound field reproducing device according to an embodiment of the present invention will be described with reference to the drawings. First, a method of localizing sound from the headphone speakers 6a and 6b to the left rear of the listener 7 will be described using the principle diagram of sound image localization in FIG.
In the figure, a signal input means 1 is for inputting an audio signal S (t), the output of which is a FIR filter 3 through left (L) and right (R) channel A / D converters 2a, 2b.
It is input to a and 3b. The FIR filters 3a and 3b are digital filters that perform signal processing to be described later, and the respective outputs are given to the headphone speakers 6a and 6b via the D / A converters 4a and 4b and the amplifiers 5a and 5b.

Now, h1 (t) shown in FIG. 3 is a head-related transfer function at the position of the headphone speaker 6a and the left ear of the listener 7. To be precise, the head related transfer function is the response at the position of the eardrum when an impulse is input to the headphone speaker.
Hereinafter, the transfer function will be referred to as an impulse response in order to explain in the time domain. However, the same result can be obtained even when considered in the frequency domain.

Similarly, h2 (t) is the headphone speaker 6b.
And the impulse response at the position of the right ear of the listener 7. Further, the signal generating means 8 uses a signal S such as an impulse signal.
is a means for generating (t), and its output is a virtual speaker 9
Given to. The virtual speaker 9 is a virtual speaker arranged on the left rear side of the listener 7. In addition, FIR filter 3
a and 3b are the impulse responses hLL (t),
It is a digital filter that performs a convolution operation with hRR (t).

Here, consider a case where a sound image is virtually localized in an arbitrary direction. Left channel headphone speaker 6a and right channel headphone speaker 6b
The principle of sound field signal reproduction for virtually generating a sound corresponding to the virtual speaker 9 provided at the left rear will be described using. First, the audio signal S (t) corresponding to the sound source on the left rear side is input to the A / D converters 2a and 2b and converted into digital signals. The digital audio signal converted here is FIR
It is assumed that the data is input to the filters 3a and 3b.

On the other hand, the signal S (t) of the signal generating means 8
When the sound is output from the virtual speaker 9, the sound reaching the ear of the listener 7 is expressed by the following equation. That is, the sound pressure L (t) at the left ear is expressed by the following equation (1). L (t) = S (t) * h3 (t) (1) The sound pressure R (t) in the right ear is expressed by the equation (2). R (t) = S (t) * h4 (t) (2) where * represents a convolution operation. Actually, the transfer function of the speaker itself is multiplied, but this is ignored. Further, it may be considered that the transfer function of a speaker or the like is included in h3 (t) and h4 (t).

Further, the sound pressure, the impulse response, and the signal S (t) in the equations (1) and (2) are considered as discrete digital signals on the time axis, and are respectively expressed by the following equations (3) to (7). To convert. L (t) → L (n) ・ ・ ・ (3) R (t) → R (n) ・ ・ ・ (4) h3 (t) → h3 (n) ・ ・ ・ (5) h4 (t) → h4 (n) ・ ・ ・ (6) S (t) → S (n) ・ ・ ・ (7)

In this case, the above equations (1) and (2) can be rewritten as the following equations (8) and (9).

[Equation 1]

[Equation 2] Here, strictly speaking, the natural number n should be expressed as nT,
Although T represents a sampling period, T is generally omitted and expressed as in equations (8) and (9). N is the length of the impulse responses h3 (n) and h4 (n).

Similarly, the signal S (t) is the signal input means 1
The following expression holds for the sound that is output as sound from the headphone speakers 6a and 6b via the and reaches the listener 7. That is, the sound pressure of the left ear is given by the following expression (10). L ′ (n) = S (n) * hLL (n) * h1 (n) (10) The sound pressure of the right ear is given by the following equation (11). R '(n) = S (n) * hRR (n) * h2 (n) (11)

If it is assumed that the sounds are generally heard from the same direction if the head-related transfer functions are equal to each other (this assumption is generally correct), the following (12) to (15) are applied.
The formula holds. L (n) = L '(n) ... (12) h3 (n) = hLL (n) * h1 (n) ... (13) R (n) = R' (n) ... ( 14) h4 (n) = hRR (n) * h2 (n) (15) Therefore, the impulse responses hLL (n) and hRR (n) are determined so that equations (13) and (15) are satisfied. Just do it.

For example, impulse responses h1 (n) to h4 (n),
When hLL (n) and hRR (n) are rewritten in the frequency domain expression, the following expressions (16) to (21) are obtained. H1 (n) = FFT (h1 (n)) ... (16) H2 (n) = FFT (h2 (n)) ... (17) H3 (n) = FFT (h3 (n)) -(18) H4 (n) = FFT (h4 (n)) ... (19) HLL (n) = FFT (hLL (n)) ... (20) HRR (n) = FFT (hRR (n )) (21) where FFT () represents a Fourier transformed (FFT) function.

Next, if the expressions (13) and (15) are rewritten in the frequency domain expression, the convolution operation is changed to multiplication as shown in the following expressions (22) and (23), and then the impulse response of each is obtained. Is the Fourier-transformed transfer function. H3 (n) = HLL (n) · H1 (n) ・ ・ ・ (22) H4 (n) = HRR (n) ・ H2 (n) ・ ・ ・ (23) In formulas (22) and (23), Since the values other than the transfer function HLL (n) and HRR (n) of the FIR filter are obtained by measurement, the transfer function HLL (n) and HRR of the FIR filter are expressed by the following equations (24) and (25). (n) can be obtained.

(Equation 3)

[Equation 4]

The transfer function HLL determined in this way
hLL that is the inverse Fourier transform (IFET) of (n) and HRR (n)
Using (n) and hRR (n), the signal S (n) is calculated by the FIR filters 3a and 3b shown in FIG. Therefore, the signal input to the headphone speaker 6a is convolved with the impulse response hLL (n), and the signal input to the headphone speaker 6b is convolved with the impulse response hRR (n). When the sound is output from the headphone speakers 6a and 6b in this way, the listener 7 does not have to actually play the virtual speaker 9 on the left rear side.
You can feel the sound coming from that direction. With such a method, the sound image can be virtually localized in any direction.

Next, a sound field reproducing apparatus according to the first embodiment of the present invention will be described with reference to FIGS. 1 is a block diagram showing the overall configuration of the sound field reproducing apparatus according to the first embodiment, and FIG. 2 is a block diagram showing the internal configuration of the signal processing means 3. In FIG. 1, the signal input means 1 is a circuit for inputting an analog (monaural) audio signal S (t) and outputting it to the A / D converter 2. The A / D converter 2 is a circuit that samples an analog voice signal at a frequency fs and converts it into a digital voice signal. The sampling frequency fs here is often set to 44.1 kHz.

The output of the A / D converter 2 is given to the signal processing means 3. The control signal input terminal of the signal processing means 3 is connected to the input means 10 via the parameter control means 11. The input means 10 is means for inputting the head shape and size of the listener 7 who is the user of the sound field reproducing device, the position of the listener 7 and the sound image, the spread of the sound field, and the like. The parameter control means 11 controls the value of the parameter (convolution coefficient) in the signal processing means 3 according to the condition set by the input means 10. The parameter control means 11 stores in advance a convolution coefficient for localizing a sound image in all directions and positions as seen from the listener, as a parameter, and selects it by the signal of the input means 10 and outputs it to the signal processing means 3. To do.

As shown in FIG. 2, the signal processing means 3 comprises left and right channel signal processing means 3c and 3d, and the left channel signal processing means 3c comprises a sound image localization means 12a, an oversampling means 14a, a filter means 15a and a down means. It is composed of the sample means 16a. The right channel signal processing means 3d is composed of a sound image localization means 12b, an oversampling means 14b, a filtering means 15b and a downsampling means 16b. Each tap coefficient (convolution coefficient) output from the parameter control means 11 of FIG. 1 is given to the sound image localization means 12a and 12b.

The sound image localization means 12a and 12b are FIs shown in FIG.
It is composed of digital filters corresponding to the R filters 3a and 3b. The sound image localization means 12a and 12b of FIG. 2 receive the audio signal sampled at the frequency fs from the A / D converter 2 of FIG. 1 and perform the calculation with the convolution coefficient corresponding to the position of the sound image. At this time, the sound image localization means 12a, 12b
Converts a parameter given from the parameter control means 11 into a time-series signal in correspondence with a delay circuit forming a part of the FIR filter and each multiplier to which the output thereof is inputted. Here, the parameters arranged in time series are given to the oversampling means 14a and 14b, and the parameters are set to n.
Oversampling at a frequency of xfs. The oversampled parameters are then filtered by the filter means 1.
5a, 15b.

The filter means 15a and 15b are low-pass filters having a cutoff frequency of fs / 2 or less. The down sampling means 16a and 16b are filter means 15
This is a circuit for inputting signals a and 15b, respectively, and sampling at a frequency of m × fs (m> 1). The outputs of the down-sampling means 16a and 16b are again converted from time-series signals into respective parameters, and the sound image localization means 12a and 12b.
Is input to The sound image localization means 12a and 12b are
When individual difference data is given to the standard head shape of the listener, which is input from the input means 10 via the parameter control means 11, calculation is performed using the corrected convolution coefficient, and the output thereof is shown respectively. 1 to the D / A converters 4a and 4b.

The operation of the sound field reproducing device configured as described above will be described. First, a voice (audio) signal is input to the signal input means 1 of FIG. This audio signal is A /
The signal is converted into a digital signal by the D converter 2 and is processed by the signal processing means 3.
Is input to The input unit 10 inputs the shape and size of the listener's head, the position and distance between the listener and the sound image, the width of the sound field, and the like. The parameter control means 11 selects a parameter so as to obtain a characteristic according to the contents input by the input means 10. That is, the parameter control means 11 selects a convolution coefficient stored in advance and outputs it to the signal processing means 3.

When the listener 7 in a certain sound field hears a sound in a certain direction as shown in FIG. 4, the impulse response input to both ears becomes as shown in FIG. 5 (a), and the frequency characteristic is as shown in FIG. It becomes like 5 (b). Then, the volume balance, the arrival time, and the phase of the left and right ears change depending on the head shape and size of the listener and the positional relationship between the listener and the sound image. First,
FIG. 1 shows the head shape of the listener 7 and the positional relationship between the listener and the sound image.
It is set by the input means 10. The parameter control means 11 selects the convolution coefficient stored in advance according to the input sound image position, and the sound image localization means 12a of FIG.
Give to 12b. The coefficient output at this time is a general value that does not take into account individual head differences. The sound image localization means 12a and 12b perform the calculation with the transfer functions shown in the equations (24) and (25). At this time, the oversampling means 14a and 14b perform oversampling in order to shift the coefficient time axis based on the difference between the standard head shape and the listener's head shape.

FIG. 6 is a general explanatory view showing the concept from oversampling to downsampling. Figure 6 (a)
Is a waveform when the input control signal (parameter) is converted into a time series signal and sampled at the sampling frequency fs. In order to oversample the signal of (a), the sampling period of the signal of (a) is further divided into n equal parts, as shown in FIG. 6 (b). Thus obtained 1
When signals are interpolated at a cycle of / (n × fs),
As shown in (c), the signals have fine intervals.

This signal is converted into the filter means 15a, 1 shown in FIG.
5b, and the cutoff frequency fs / is added to the interpolated signal.
When input to the low pass filter No. 2, the signal is sampled and held, and a signal having a smooth curve as shown in FIG. 6D is output. This signal is downsampled by 16a, 16
When given to b, sampling is performed at a frequency of m × fs (n>m> 1), and a signal as shown in FIG. 6E is generated. In the signal shown in FIG. 6E, the number of data is increased 14/11 times at the same time interval as compared with the signal shown in FIG. Thus, a time-series control signal finer than the original sampling waveform is obtained. Then, this time-series signal is given to, for example, a serial-in / parallel-out shift register, and when the respective signals are simultaneously taken out, a control signal corrected to a desired value is obtained.

Next, applying this method to the convolution coefficient,
A case of realizing more accurate sound image localization will be described.
FIG. 7 shows the concept of how to shift the data with respect to the individual difference of the convolution coefficient. FIG. 7A shows a typical convolution coefficient statistically calculated from many listeners with respect to the input voice signal. Therefore, if the sound image localization means 12a and 12b are used for convolution using this coefficient, a considerable listener should be able to perceive a natural sound image localization feeling. However, in many cases, the shape of the head varies depending on the listener, and is large or small. Therefore, the standard coefficient causes a slight difference in the binaural distance, which requires correction for each listener.

As this correction method, a method is used in which the coefficients are temporally shifted to change the phase difference between the sounds reaching the left and right ears. If the coefficient of FIG. 7A is shifted by one tap, the result becomes as shown in FIG. 7B. For example, sampling frequency fs is 44.1kHz
When sampling at this frequency, the binaural gap difference is limited to 7.7 mm even if the coefficient is shifted by one tap.
Therefore, it is not possible to deal with individual differences with binaural distances of 7.7 mm or less. However, when oversampling n times, 7.
It will be possible to handle individual differences of about 7 / n (mm).

FIG. 7C is an oversampled coefficient shown in FIG. By shifting by 1 tap in this state, the phase can be finely shifted even with the same coefficient as shown in FIGS. 7D to 7F. In particular, it is an effective method in the case where a delicate binaural difference of an individual greatly affects the sense of sound localization, such as in the case of an inner phone. In this way, the sound image can be localized in a desired direction for any listener.

Next, a sound field reproducing apparatus according to the second embodiment (claim 3) of the present invention will be described with reference to FIG. FIG. 8 is a block diagram showing a part of the configuration of the sound field reproducing apparatus of the present embodiment, and shows only the circuit added to the subsequent stage of the signal processing means 3c, 3d shown in FIG. The other blocks are the same as those in the first embodiment, and the description of the entire configuration is omitted. As shown in FIG. 8, reverberation sound producing means 17a and 17b are provided for the output signals of the signal processing means 3c and 3d.

The reverberation sound producing means 17a and 17b are means for producing and adding reverberation sounds according to the width of the sound field, and the outputs thereof are given to the D / A converters 4a and 4b of FIG. 1, respectively. The reverberation sound creating means 17a and 17b are configured by, for example, connecting a plurality of feedback echo circuits having different delay times in series. For example, when the reverberation sound has a spatial spread of about 10 m, the length of the reverberation sound is, for example, 0.25 to 0.35 seconds. Further, when the space is widened to a sense of distance of 10 m or more and 20 m or so, the length of the reverberation sound is set to 0.7 to 0.9 seconds, for example. Also, if the sound field is reproduced like a large concert hall, the reverberation time is relatively long,
In addition, the reverberation time in the low range is added to be longer than that in the high range.

FIG. 9 is a conceptual diagram of parameter control of the sound field reproducing apparatus in the third embodiment (claim 4). The configuration of the sound field reproducing device of this embodiment is also the same as that of FIG. In this embodiment, the position of the sound source and the width of the sound field are input by the input means 10, and the lateral reflection angle θ indicating the distance relationship between the listener P and the sound image S is input. For example, as shown by the listener P2 in FIG. 9, as the distance from the sound image S increases, the value of θ decreases. Further, as the listener P1 becomes closer, the value of θ becomes larger. It is assumed that the delay and the convolution coefficient in the signal processing means 3 change depending on the magnitude of the side reflection angle θ.

FIG. 10 shows a fourth embodiment of the present invention (claim 5).
2 is a block diagram of the sound field reproducing apparatus in FIG. 3, and the same parts as those in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted. As shown in this figure, a parameter receiving means 18 is provided in place of the input means 10 of FIG. The parameter receiving means 18 is means for receiving a control signal for controlling the position, distance, and sense of spread from the outside.

When the parameter receiving means 18 receives a control signal including conditions such as a position, a distance and a feeling of spread from the outside, the parameter receiving means 18 gives the control signal to the parameter control means 11.
Then, the parameter control means 11 outputs parameters (each coefficient) to be set by the signal processing means 3. The subsequent operation is similar to that of the first embodiment described above.

As described above, in each of the embodiments, it is possible to finely adjust the phase by over-sampling the convolution coefficient for preset sound image localization for any listener. Then, the sense of localization, the sense of distance and the sense of spaciousness of the sound can be controlled by a control signal from the outside. It is possible to control the sound localization, the distance, and the spaciousness according to the.

In this embodiment, the case where the input signal is a monaural audio signal has been described. However, even in the case of stereo, it is possible to cope with this by providing an independent signal processing means for each signal, and instead of headphones or inner headphones. It is also effectively applied to a 2-channel speaker.

[0041]

As described above, according to the present invention, parameters are set according to the positions of the listener and the sound image input to the input means, the shape and size of the listener's head, the distance and the width of the sound field. Control and over-sampling the convolution coefficient for localizing the sound image, it is possible to deal with the delicate binaural difference of the individual for any listener, and natural sound image localization and sound field reproduction can be realized.

[Brief description of drawings]

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

FIG. 2 is a block diagram showing a configuration of a signal processing means used in the sound field reproducing apparatus of the first embodiment.

FIG. 3 is an explanatory diagram showing the principle of sound image localization in the sound field reproducing device of the present invention.

FIG. 4 is an explanatory diagram showing a relationship between a listener and a sound image.

FIG. 5 is a graph showing frequency characteristics and impulse response of both ears.

FIG. 6 is a conceptual diagram showing an operation principle from oversampling to downsampling.

FIG. 7 is a conceptual diagram of how to shift the convolution coefficient with respect to individual differences of listeners.

FIG. 8 is a block diagram showing a partial configuration of a sound field reproducing apparatus according to a second embodiment of the present invention.

FIG. 9 is a conceptual diagram of parameter control in the sound field reproducing apparatus of the third embodiment of the present invention.

FIG. 10 is a block diagram showing a configuration of a sound field reproducing device according to a fourth embodiment of the present invention.

FIG. 11 is a block diagram showing a configuration example of a conventional sound image localization device.

[Explanation of symbols]

 1 signal input means 2, 2a, 2b, A / D converter 3, 3c, 3d signal processing means 3a, 3b FIR filter 4a, 4b D / A converter 5a, 5b amplifier 6a, 6b speaker 7 listener 8 signal generation Means 9 Virtual speaker 10 Input means 11 Parameter control means 12a, 12b Sound image localization means 14a, 14b Oversampling means 15a, 15b Filter means 16a, 16b Downsampling means 17a, 17b Reverberation sound creating means 18 Parameter receiving means

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

Claims (5)

[Claims] 1. An A / D converter for converting an input audio signal into a digital signal, and the digital signal converted by the A / D converter are input, and the head shape and size of the listener, the listener and Signal processing means for performing signal processing so as to obtain an acoustic signal corresponding to the position and distance of the sound image; input means for inputting the shape and size of the listener's head and the position of the listener and the sound image; and the input means Parameter setting means for setting the parameters so that the acoustic characteristics corresponding to the input contents are obtained, and giving the parameters to the signal processing means, and the signal output from the signal processing means for converting into an analog signal D / A converter, an amplifier that amplifies the output of the D / A converter, left and right acoustic converters that acoustically convert the output of the amplifier,
A sound field reproducing apparatus comprising: the signal processing means, according to a positional relationship between the listener and the sound image input by the input means, a sound image localization means that localizes a sound image by convolution calculation, and the sound image The standard convolution coefficient is input as parameters from the parameter control means via the localization means, these parameters are converted into a time-series discretized signal, and the A / D
Oversampling means for oversampling the parameter at a value equal to or higher than the sampling frequency of the converter; filter means for passing a component having a predetermined frequency characteristic or less with respect to the parameter processed by the oversampling means; The parameters output from the above are downsampled with a value equal to or lower than the sampling frequency of the oversampling means, the obtained time-series signals are converted into convolution coefficients in parallel, and the values are used as corrected convolution coefficients. And a down-sampling means for giving to the sound image localization means. 2. The sound field reproducing apparatus according to claim 1, wherein the filter means passes a frequency component of ½ or less of a sampling frequency of the A / D converter. 3. The input means inputs the head shape and size of the listener, the position of the listener and the sound image, and the spread of the sound field at the same time. The output of the signal processing means 2. The sound field reproducing apparatus according to claim 1, wherein the section is provided with reverberation sound creating means for adding reverberation sound corresponding to the width of the sound field obtained from the input means. 4. The parameter control means, when the sound source, the reflected sound direction, and the angle viewed from the listener are input in the reflected sound that is reflected by the wall from the sound source in the acoustic space and is incident on the listener, 2. The sound field reproducing apparatus according to claim 1, wherein the angle is converted into a listening position or a sound image position in the acoustic space and a parameter is given to the signal processing means. 5. The sound field reproducing apparatus according to claim 1, wherein the input unit includes a parameter receiving unit that receives a control signal relating to a sound field from the outside and supplies the control signal to the parameter control unit.