JPS61195099A - sound reproduction device - Google Patents

Abstract

PURPOSE:To obtain most suitable sound reproduction system for accoustic sensitivity by calculating a summation coefficient of a convolver taking the correction of accoustic sensitivity into account from viewpoint of actual stage. CONSTITUTION:At listening position 8, a convolver 5a necessitates the summation coefficient which minimizes square low of difference between an input signal of an L channel from an input terminal 4a to the convolver 5a and inpulse response obtained at listening position 8. For a convolver 5b, the summation coefficient is obtained in the same way. Further this calculation must be made by convoltional integration. Then transfer characteristic is changed by calculation control of each summation coefficient. In this case, when specified characteristic in the listening position 8 is set to be an ideal inpulse, if the ideal inpulse position, where the square low of difference becomes minimum value p, exceeds a delay time of 50 msec or more, the ideal inpulse position must be set forward within the limit of two times of Minimum value p.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an audio reproducing apparatus that corrects an audio transmission system in a reproducing system including an acoustic space of an audio signal.

2. Description of the Related Art In general, the acoustic transmission characteristics of an audio signal in a reproduction room, including the characteristics of the speaker system, are influenced by the reflected sound of the room. There is a method of correcting the amplitude frequency characteristics as a way to flatten the sound reproduction system at the listening position, but since the phase frequency characteristics are not corrected, it is not possible to sufficiently improve the sound quality, sound image quality, etc. in stereo playback. do not have. In order to overcome this drawback, an application has been filed in which a convolution integrator circuit is provided in the audio signal transmission path, and the transmission characteristics are varied by calculating and controlling the weighting coefficient of this convolution integrator circuit. (Japanese Unexamined Patent Publication No. 5S-SOA1'; No. Z, Utility Model Application No. 59-Sho.
10628E1).

FIG. 7 shows a block diagram of the conventional sound reproduction device mentioned above. In FIG. 7, 1 is a delay circuit having a large number of output terminals, 21 to 2n are multipliers that control weighting coefficients for a plurality of outputs from the delay circuit 1, and 3 is an adder. A convolution and integration circuit (hereinafter referred to as a convolver) 6 is configured to vary the transmission characteristics of the audio signal applied to the terminal 4. 6 is a power amplifier, 7
is a speaker.

In the transmission circuit having such a configuration, in order to obtain a weighting coefficient that minimizes the square of the difference between the impulse responses obtained at the input terminal 4 of the convolver 5 and the listening position 8 at the listening position 8, the multipliers 2a to 2 are used. It is configured to change the value of 2n. In such a state, if the discrete coefficients of the impulse response of the convolver 6 are gl, g2...gm-1, then
Discrete response f0. at listening position 8. f1...'n
+m-2 is expressed by the following formula.

However, hi is the transfer characteristic p=o...-...n+m-2.

Expressing equation (1) as a matrix, we get ・・・・・・・・・(barrel) Then, The @) formula can be further expressed as F=)IG.

Here, if we take the square of the difference between the input impulse F0 and the impulse response F at listening position 8 and set it as an evaluation function p, then p=(y-yo)T(y-yo)=(HG-yo)T (HG-Fo) = (GTHT-FOT) (HG-70) = GTHT
HG-FOTHG-GTHTFO+F”FO,,,,,
-, , (3), and when the innopulse response G of the convolver 6 is determined so that the evaluation function P becomes the minimum, the following shows the transposed matrix.

Then, from equation (4), HTHG = HT Fo...
・・・It becomes an individual.

That is, by setting the coefficients of the multipliers 2a to 2n as shown in Equation 5 above, the transmission characteristics are changed, the sound pressure frequency characteristics at the listening position are flattened, and the original sound quality can be reproduced with high fidelity.

When calculating the weighting coefficient of the convolver using the above conventional method, if the target impulse Fo is set to the NBth
is F0= Co , o , −=−, 1, o , =−・
o )・・・・・・・・・・・・・・・(6) Note that "1" in formula (6) is the NBth one.

The evaluation function P is influenced by the order in which this target impulse is set. Conventionally, the target impulse IF is located at the position where the evaluation function P is minimum.

was set.

Problems to be Solved by the Invention However, in the conventional configuration described above, when the target impulse Fo is set at a position that minimizes the error, when convolution with the audio signal is performed, the position of the target impulse, that is, the delay time, changes the auditory sense. Upper echoes may be felt, causing problems in sound reproduction. Also 2CH
In the case of stereo reproduction, the position of the target impulse that minimizes the error does not match.

The present invention has been made in view of this point, and the evaluation function P
The objective is to provide the best sound reproduction system in terms of audibility, taking into account not only the minimum value of , but also the target impulse position, that is, the delay time.

Means for Solving the Problems The present invention aims to solve the above problems by using the evaluation function P determined by equation (3) and the position (delay time) of the target impulse.
From the relationship, the delay time of the minimum error value is 5QIIIS
When the delay time exceeds eC, the delay time is set within 50m5eC at the expense of the minimum value of the error, thereby improving the auditory sense from both characteristics of the frequency 9 time axis. In the case of 2CH reproduction, the average of both channels is taken and the delay time is set in the same manner as above.

Effect By adopting the convolver design method described above, the flatness of the sound pressure frequency characteristics at the listening position, including the playback room, can be designed not only from the time axis but also from the perspective of auditory sensation. This is also an effective method from a practical point of view for sound reproduction.

EXAMPLE An example of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a sound reproducing apparatus according to an embodiment of the present invention, in which the same parts as in FIGS. 2 and 7 are given the same numbers.

In Figure 1, 4! L and 4b are R for audio signals, etc.
Channel (hereinafter referred to as CH), I, input terminal for CH input signal, 5! L, 5b is a convolution integrating circuit (hereinafter referred to as a convolver) that varies the transmission characteristics of the input signal, e
&, eb is a power amplifier, and a.

7b is a speaker, 8 is a listening position for speakers 7a and 7b, and 9! L and 9b are preamplifiers that amplify input signals to the convolvers 51L and 5b, and 1o is a regeneration chamber.

Regarding the present embodiment configured as described above, the operation will be described below only for 1CH for convenience.

First, at the listening position 8, the L channel input signal obtained from the input terminal 4a to the convolver 6 &
The weighting coefficient that minimizes the square of the difference between the impulse responses obtained by is determined using the four holes described above.

Since the number of delays of this Tokikon Bolpa 6a is finite,
The error varies depending on the number of delays. This situation is shown in FIG.

In FIG. 3, the horizontal axis represents the position of the target impulse (that is, the delay time t) can be expressed by the following equation (7).

t=ΔtXNB ・・・・・・River...
(7) Note that Δt is the sampling interval and NB is the position of the target impulse. ), and the vertical axis is the evaluation function P. It can be seen that as the number of delayed volumes of the convolver 6a increases, that is, the number of taps, the error in the evaluation function becomes smaller and the tolerance range becomes wider. Depending on the relationship between the reverberation time of the reproducing room 1°, the number of delayed volumes of the convolver 5a, and the control frequency band, the configuration shown in FIG. 1 may be able to correct only the low frequency range.

Now, set the sampling frequency to 73. If the number of books delayed by the convolver is n, the controllable time length T is as follows (8)
The formula becomes

T=n/Fs ・・・・・・・・・
(8) It can be seen that the larger the number of books delayed by the convolver and the lower the sampling frequency, the longer the control period can be achieved.

Generally, the number n of books delayed by a convolver is often limited by the scale of the hardware. In Fig. 3, according to the conventional design method, NB = 2oO, and in order to lengthen the reflected sound control time length, if the sampling frequency 'FB is set to 3, the position where the impulse is generated is (from the η formula, t = 1
/3000×200=66 (11SeC). If the configuration shown in FIG. 1 is adopted under these conditions, the reproduction band will be 1-5 KHz from the sampling theorem. Therefore, in another embodiment shown in FIG. 2, the low-pass filters 11&, 11b
and bypass filter 13a.

13b performs band division. low pass filter 1
The input signal of 1.5K) 12 or less that has passed through 11L is corrected by the convolver 6a.

On the other hand, the signal passing through the bypass filter 13a is delayed by the delay time of the low frequency band from the delay line/14a, and the delay times of the low frequency band and the high frequency band are combined and added by adders 12L and 12b, and then reproduced.

Now, in FIG. 1, the impulse response and frequency characteristics at the listening position 8 when the convolver 5a is not passed are shown in FIG. Next, FIG. 6 shows the impulse response and frequency characteristics obtained by the conventional convolver design method. The position of the target impulse at this time is NB=2oo, and P=0.043
It is. The delay time is 66 m, and the frequency characteristics are flattened to within ±1° sdB from 26 frequencies to 1.5 KHz. However, the delay time of 66 Bsec is audibly felt as an echo when convolution and integration are performed on the music signal.

Therefore, in this embodiment, the position of the target impulse is moved forward at the expense of some waveform error. In this case, the value of the evaluation function P is limited to about twice the minimum value. P==
1.0. FIG. 4 shows the impulse response and frequency characteristics when NB=1oo.

Looking at FIG. 4, the degree of approximation of impulses and the flatness of frequency characteristics are worse than in FIG. 6. However, since the delay time is set within som5ec, which is not perceived as an echo by the human auditory sense, the characteristics shown in FIG. 4 are good for the sound reproduction system. The above has been described for 1CH, but in the case of 2CH, the average of both OH is taken and the same method is used.

Effects of the Invention As described above, according to the present invention, in addition to the conventional design method, the weighting coefficient of the convolver is calculated by incorporating auditory correction from a practical standpoint, thereby creating a listening position that does not cause any auditory problems. This achieves flat characteristics at .

[Brief explanation of drawings]

Figure 1 is a block diagram of a sound reproduction device according to one embodiment of the present invention, Figure 2 is a block diagram of another embodiment of the present invention, and Figure 3 is a convolver weighting coefficient calculation method using the number of delayed books as a parameter. A characteristic diagram showing the relationship between delay time and evaluation function, FIG. 4 (IL) is an impulse response characteristic diagram when the weighting coefficient of the convolver is calculated based on the present invention, and FIG. 4 (b) is the same frequency characteristic diagram. , Figure 5 (IL) is an impulse response characteristic diagram at the listening position when no convolver is passed through, Figure 5 middle) is the same frequency characteristic diagram, Figure 6 (!) is the same frequency characteristic diagram.
L) is an impulse response characteristic diagram obtained by a conventional convolver weighting coefficient calculation method, FIG. 6(b) is a frequency characteristic diagram thereof, and FIG. 7 is a block diagram of a conventional sound reproduction device. 4a, 4b... Input terminal, 5N, 5b...
...Convolver, 6a, 6b...Power amplifier, 7! L, 7b...Speaker, 8...
...Listening position, 9&, 9b, River...Preamplifier, 1o
・・・・・・Reproduction room. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 3
Zutenda me'M.・t・Nβ] [“8] Fig. 4 (sil Fig. 5 [HK] Fig. 6 IVI [Hz] Fig. 7

Claims (2)

[Claims] (1) A convolution integration circuit is provided in the audio signal transmission system, and the impulse response with specific characteristics at the listening position is
Audio signal transmission configured to calculate and control the weighting coefficient of the convolutional integration circuit by calculating the square of the difference from the listening position impulse response by setting the weighting coefficient of the convolutional integration circuit, thereby changing the transmission characteristics. If the position of the ideal impulse that has the minimum value P of the square of the difference has a delay time of 50 msec or more, the limit is twice the minimum value P. A sound reproduction device characterized in that the ideal impulse position is set forward. (2) The sound reproduction device according to claim (1), wherein in a transmission system for audio signals of 2 channels or more, the position of the ideal impulse is set forward based on the average value thereof.