JPS58135921A - Delay time detector and acoustic device - Google Patents

Abstract

PURPOSE:To improve preference (comfort on hearing sensation) in an audio device such as hearing aids and to facilitate hearing, by finding a delay time of a reflecting sound or a residual sound time, which is felt comfort from the viewpoint of a hearing sensation of a human being, from a self-correlative function of a sound source. CONSTITUTION:An audio signal, applied to an input terminal 1, is transferred to multipliers 25-28 and a delay circuit 22. A plural of different delay outputs D1...DN-1 are obtained by delay circuits 22-24', and in the multipliers 25-28, a multiplication takes place between input signals or an input signal and each delay output. In integrating circuits 29-32, values thereof are respectively integrated, and an absolute value is found by absolute valuing circuits 33-36. Outputs phio...phiN-1 of the absolute valuing circuits 33-36 corresponds to an absolute value of a self-correlative function of an inputted audio signal. After an absolute valuing outputs phio of a signal, conducting no time delay, is damped to 1/10 or 1/4 by an attenuater 37, it is compared with other absolute valuing output by comparing circuits 38-40. An absolute valuing output, being smaller than the output Ref of the attenuater for a signal conducting no time delay, is selected by the comparing circuits 38-40, and a comparing output, related to a lowest or highest delay output of a delay time, is selectively coded by an encoder 41. Namely, a delay output, corresponding to an adding means having an adding output being nearest to the damping output Ref, is selected.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a delay time detector and a children's rhyme device for detecting the optimum delay time of reflected sound that provides more preferable auditory characteristics.

A comparison between the human auditory psychology or sense of hearing and the physical characteristics such as the reverberation time of a room or the acoustic characteristics of the sound source itself.
ζ has an sr-tangential relationship.

There is an autocorrelation pre-del of the warp, but it is said that the delta before the warp threads also indicates the close relationship between the sense of hearing and the sound source.

FIG. 1 is a Licklider model of the auditory nervous system. In Figure 1, l is the direct nerve, ! is a delayed nervous system,
8 is an input, and 4 is an output No. 1 is a synapse. An impulse synchronized with the envelope period of the original signal, which has been subjected to some degree of frequency analysis in the pot, enters from the input field. When the first impulse enters, it is transmitted through the direct nerve L.

It also transmits delayed nervous system thoughts via synapse 6.

When the second impulse enters, the signal on direct nerve 1 catches up with the delayed first impulse.

The simultaneous occurrence of these impulses causes output 4 to be triggered by 1.
Time-place ib takes place. In other words, the human auditory nervous system has an autocorrelation function.

Recent research on hearing has revealed that reflected sounds □ and subsequent reflected sounds that arrive later than direct sounds have a heavy influence on the sense of hearing.

Using music and speech, Ando evaluates a synthesized sound field consisting of m-single and reflected sounds when reproduced through a speaker, using a scale of preference (comfort over the human body). According to this conclusion, the normalized autocorrelation function ψ of the source signal is
(The level of one reflected sound is ±64 of the direct sound.)
It has become clear that when summer changes to 11, the optimal delay time for the reflected sound is ξ, where 1(r)1 corresponds to a time corresponding to 1/10 of the level ^ of the first reflected sound. Figure 3 is 19 of ξ, (l is 1/l0I of j11 reflected sound bell^
The horizontal axis represents the time corresponding to C (τd), and the vertical axis represents the delay time Tl111 of the single reflected sound at which the preference is maximum. The range shown in the figure shows the delay time when the preference is 0.1 lower than the maximum value, and the symbol 0 in the figure is ^■6-1.・ indicates A&■048°mouth is ^-4$41Tt. In particular, if we call the time when 1 becomes 01 times ψ, (O) re(0,1).

In the case of Almon, it can be expressed as 14 # rv (01). The %t- is the delay time of the single reflection sound where the preference is maximum! It can be seen that there is good agreement with -.

In addition, Ando et al. conducted a similar experiment in the case of monaural listening using earphones, and in this case, the maximum value of preference for a single reflected sound was 1ψ, (τ)1 was ψ,
It is clear that this can be obtained at τ5(O40), the time when the value becomes 0.26 times (0). In Figure 3, the horizontal axis represents the delay time Δt1 of a single reflected sound + the vertical axis represents the normalized preference, where (a) is speech, (b)
is the time for music. In fig.

Those with normal hearing and hearing squirrels (IIIg) are 70
~96-B, and hearing loss (IKHg
) is 10 (ml).
.

It is expressed in . As is clear from this graph, the delay time of a single reflected sound is re(O
It can be seen that the preference takes the maximum value at the time of 40). In other words, this result shows that if this is taken into consideration when designing a hearing aid, a hearing aid with characteristics that are easier to hear can be obtained.

Furthermore, it has been reported that the autocorrelation function of the sound source signal has a close relationship with the optimal persistence time. Figure 1 shows the measurement results. The horizontal axis shows the above-mentioned f@(0, IL the vertical axis shows the desirable median value of reverberation time is mouth b). , is expressed as the time it takes for the signal in the reverberation section to decay by 60dll.In the figure.

A, B and Tomi are music, S is speech, but Na・Chib]
It can be approximately approximated by the function dthI(H±10)τ@(0,1).

In view of the above-mentioned situation, the present invention provides a detector for determining the delay time or reverberation time of a reflected sound that is perceived as preferable to human hearing from the autocorrelation function of a sound source, and an acoustic device 1 applying the detector. It is an object.

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

In diagram li6 (Il), 21 is the input terminal, ! 3～
24 is a delay circuit, 2B-1111 is a multiplier, 29 to 8 are integrating circuits, II to Hatake are absolute value conversion circuits, IT is an attenuator, 88 to 40 are comparison circuits, and 41 is an encoder.

Here, the function and operation of the invention will be explained. When an acoustic signal is applied to the input terminal l, the multiplier 35~! ! 8 and delay circuit 22. Each delay circuit 22 to 24' produces a plurality of different delay outputs DI1D2e''・DN
-1 is obtained, n1 multipliers 26 to 28 perform multiplication of the input signals or the input signals and each delayed output.

Thereafter, the values are integrated in the respective integrating circuits 29-82, and the absolute values are determined by the absolute value converting circuits 88-86.

It goes without saying that the integrating circuit generally needs to be initialized before it starts integrating, and in some cases a low-pass filter can be used instead. Output #6 of absolute value conversion circuits 88 to 86 *Il h ”.

#N-1 corresponds to the absolute value of the autocorrelation function of the input sound number 1. FIG. 6 (11) shows - when the signal is obtained by removing the high M wave number component from the input using a low-pass filter. , 1. . . . #N-1, and when configured as a simple type, this level is sufficient. -0 is the maximum value and it is normal as s II'S * at...

Its value becomes smaller.

Absolute value output #0 of a signal without time delay is attenuated to 1/10 or 1/4 by attenuator 87, and then compared with other absolute value outputs by comparison circuits s8 to s40. These comparison circuits 18 to 40 select the absolute value output smaller than the attenuator output I of the signal that is not delayed by one hour, and selectively select the comparison output related to the delayed output with the smallest delay time. It is encoded by the encoder 41. still.

The comparison output related to the longest delayed output (thicker than the foot f good.

In addition, the damping ratio 01 or QJ5 mentioned above is o, l±0.8
Furthermore, even if j is within the range of QJS±0.08, almost the same effect can be obtained.

FIG. 6 shows an example of the application of the fsl constructed using the n hours a week detection system according to the present invention. this is.

It is an optimal reflected sound addition device that can be applied to 41 devices (including hearing aids) to make children more comfortable and easy to listen to. In FIG. 1, snow l to 41 constitute the delay time detection 11 in FIG. 5, and perform similar operations. The input signal τe (
An output corresponding to 0, 1) or T.(HI8)- is obtained from the encoder 41, and is stored in a latch 41ζ to hold it for a not too short time.

The optimum delay time for a single reflected sound is such that it does not need to be latched suddenly or frequently, so it is configured to latch at intervals of not too short a time. There is an odor when the latch is switched, which is audible.
The attenuator 44 is designed to gently attenuate the delayed signal so as not to give the user a sense of noise. This latch signal generating section is not shown. Depending on the output of latch 42, delay output D1. D! , . . , DH-1 is selected by an analog multiplexer 48, passed through an attenuator 44, and then added to the input signal by an adder 45 to obtain an output 46.

With such a configuration, it is possible to realize a sound stimulation device that has a large preference (aural comfort) and is easy to listen to.

#H7 shows another application example constructed using the delay time detector according to the present invention. This is an optimal residual addition device for creating FM and easy-to-use sounds that are more suitable for use in sound reproduction devices. In FIG. 7, numerals 21 to 41 constitute the slow-talking time detectors in FIG. S and FIG. 6, and perform similar operations. Signals necessary for setting the optimum residual sound time are obtained by the code converter 41 which receives the comparison signals S, . . . I-1. This signal is latched at an appropriate timing, and the black portion is similar to the case in FIG. 96.

This latched signal is a residual time setting signal, and is M
Given to four parable circuits. 4 SW circuits have input terminal 21
In response to the input signal supplied to the input signal and the above-mentioned early spring time setting signal, a residual additional signal output 46 is generated.

With such a configuration, it is possible to realize an optimal accelerator attachment device that has the highest hearing comfort and is easy to hear.

As described above, according to the present invention, it is possible to construct a hearing aid that is more preferable, has a high reference level, and has easy-to-hear sounds, and various other nursery rhyme devices.

Although the details of the residual *a path 47 were not clearly shown in the application example of m1ll in FIG. Any device may be used as long as the time can be set variably, and output signals from delay circuits 22 to 24 or the like may be used.

In the above explanation, + 10 # #l a-・-t1
As a simplified form for determining the value of , we have described the case where a signal obtained by removing high frequency components from the input signal is processed as input. The reason for this is -0, #1, -11w-1
is a monotonically decreasing function as shown in Figure 6(b).
This is because we have taken care to ensure that Similarly, to obtain the function shown in Figure 5(b), input a filter that passes only periodic components that are integral multiples of the delay time of the four delay circles! ! One example is to perform processing using the above-mentioned circuit after being provided before the circuit. In this case, although it is a little more complicated, it takes into consideration high frequency components, and the degree of approximation also increases.

Moreover, the block prisoner of the correlator shown in FIG. 6 may be changed in any way, and in FIG. ) It goes without saying that a converter may be provided at the point and the signal processing may be performed using digital data. Further, although the method for configuring the correlator of C is such that a<normally n is selected, a configuration such as Schroeder's square integral type residual device may also be used.

Next, this square integral form 8! Example 1klI8 using a musical instrument
The diagram will be explained.

In Figure 8, the input terminal! ! When an acoustic signal is given to l, it speaks. Output to integrating circuit 8 to I2 - (11#1+...#N-
1##N-1 is obtained in the same manner as in Example 111.

Integral output -o, #1. -#M-1 is each multiplier 5
Cubed by 1-64. The multiplied outputs are summed by adders 55 to 67 respectively, ie, squared and integrated, and the output of adder 67 is attenuated by attenuator 87. As this damping ratio, it is preferable to use a value within the range of 1 east value of the value in the first embodiment of Fig. S (a), that is, Oo. . After this, the comparator circuits 118 to 40 output the outputs of the eight attenuators.
The output that is smaller than QR@l compared with f and that is closest to the six adders is selectively encoded by the encoder 41. In terms of circuitry, the output of the adder 66 is larger than the output of the adder 16, and similarly, it goes without saying that the output of the six adders is the largest. In the case of this j1!2 embodiment, the signal that becomes one table for determining the optimal delay time to be obtained is the #N-1 signal obtained from the output of the integrating circuit 8g, contrary to the first embodiment. It should be noted that this is the output of the adder 67 corresponding to .

Furthermore, we will discuss a method for more precisely determining #@a #l a ''・-t1. Figure 9 shows an example of its configuration. In Figures (a) and (b), 4i1
is a delay circuit.

62 is a multiplier, 68 is an integrating circuit, and 64 is an absolute value conversion circuit.

6b is a delay circuit I11. Multiplier 62. A circuit (4) including an integrating circuit 68 and an absolute value converting circuit 64, and a circuit 66 are maximum value detecting circuits. In FIG. 9(a), 31 summer is an input terminal, Dlj is a delay input terminal, D'''oe is a delay output terminal, and -1 is a correlation output terminal. By giving SIN and Dlj, the correlation output signal l Using this circuit (A) and the configuration shown in FIG. 9(b), #0・−1・°
°° - 0 to find 1. -0 is obtained by integrating the cube value of the input signal, and it goes without saying that the absolute value converting circuit 8s is not necessarily necessary. Next, the circuit (4
) 0 shares in l.

I1) Given l, we get -1. Similarly, −ffi+
11g, .'.-0 are obtained and 1. For example, in the case of this embodiment, the maximum value is selected from among -1 of one eyelid by the print large value detection circuit 66 and set as a' fl. Hereinafter, #, , #, , . . . -N-1 will be obtained by the same method. Normally, there is no need to calculate, and since -0 is the largest, the circuit for deriving -1 to -1 and the maximum value detection circuit -6 for determining #0 can be omitted. As mentioned in Figure 5(b), if we calculate 00,11g''-tl for p, even if the acoustic input signal contains an island frequency component, its value will be as a function that decreases almost monotonically as shown in Figure 5(b). can be obtained, and the optimal delay n time can be determined well.In the case of jin, l&
A predetermined delay time is selected from among the delay times corresponding to the output of the maximum value detection circuit closest to the variable delay zero time R@f.

FIG. 10 shows a further improvement of the embodiment shown in FIG.

In the figure, 70 is a comparator, and 71 is an AND gate.

The case of jin is different from the case of the example in Figure I9.

Instead of finding #, , l, , -#N-1 (, it is possible to find the input S1*1!*...5N-1 of the encoder 41. Obtained similarly, below -1°-1,..., -1° (N-1)
The value of can also be found in almost the same way.

Signal -0 is attenuated by attenuator B7 to create if.

-1,φ1. ..., φ,. (N-1) by the comparator τ0. The comparator 70 is R@f↓rimo-1,φ*
* "= # 1 It is configured so that the output is at a high level (hereinafter abbreviated as "I") when n(N-1) is smaller. The signal S,,!l,, ・-, 8N
−1 is −1 to −10+ −11″−1(l
a"", -1o(jl-2)+1~-to(N-1)
is the output of AND game 1, where each input is -1
～-1o, -11～-addition, -to(father-in-law-2)+1～-s
When all of o (N-1) become smaller than the foot f, it becomes °'H". This is encoded by encoder 4.
1 and outputs the delay time of a predetermined delay circuit corresponding to the AND game 1 to which the delay circuit having the longest delay time among the AND games 1 belongs.

With this configuration, the AND game 1 is necessary, but the maximum value detection circuit 6 is unnecessary, and it is advantageous to have one ghost in one circuit configuration. In both examples, 10 was calculated in advance.
1 from Kenono value (D ## (#-0, N-1)
, Sm (mm1, N-1), but it goes without saying that the number is not necessarily 10 and may be any other number.

[Brief explanation of drawings]

Figure 1 is a diagram showing a model of the human auditory system based on beam Tsurider, Figure I2 is a diagram showing the relationship between τ- and the late n time of a single reflected sound at which the breference is maximum, and Figure 8 is a diagram showing the Figure 4 shows the relationship between the reflected sound delay time Δt1 and the normalized Samuta preference. , Fig. 6 shows the optimum reflection time.A block diagram and an operation explanatory diagram showing an embodiment of the delay time detector according to the present invention for obtaining the information ma or rd for obtaining the reverberation time. FIG. 7 is a block diagram showing an optimal first reflected sound adding device as an application example of lil using the delay time detector according to the invention, and FIG. FIG. 8 is a block diagram showing a third embodiment of the present invention; FIG. 189 is a block diagram showing a partial configuration and a block diagram showing an embodiment of the present invention; FIG. 10 is a block diagram showing an embodiment of lI4 of the present invention. l...direct nerve, 2...delay nervous system, heat...input, 4--output Is synapse, gl--input terminal,
2 to 24-・delay circuit, 26 to 24-・multiplier,! 9
~8! -Integrator circuit. 8 to I6-absolute value circuit, 87--attenuator, lll-
40-・Comparator j11.41...Encoder or code converter, 4! -Latch, 48...Multiplexer, 44...Second attenuator, 4I-・Adder, 4L-・
Output, 4 ma...reverberation circuit, I1...delay circuit, 6 to...multiplier, 5sst integration circuit, 64...absolute value circuit, 6 to...circuit (^), -6...・Maximum value detection circuit, TO-comparator, 71...AND gate. Note that the same reference numerals in Figure 1 indicate the same or equivalent parts. Fig. 5 (L) Fig. 5 (b) 9 f, +2 s, y fs-z fs- Figure 10

Claims (1)

[Scope of Claims] (1) An input terminal to which an acoustic input signal is applied, a delay means for delaying the input signal and generating a plurality of delayed outputs with different delay times, and the input signals are mutually or the input signal and the delay output are Multiplying means for generating a plurality of multiplication outputs by multiplying each of the multiplication outputs respectively; Integration means for generating a plurality of integral outputs by integrating each of the above multiplication outputs 1 Converting each of the above integral outputs into an absolute value and converting them into a plurality of absolute values Absolute value converting means for generating an output; attenuating means for generating an attenuated output by attenuating an absolute value output related to the multiplication output of the input signals among the absolute value outputs at a predetermined attenuation ratio; and the above absolute value. Comparing means generates comparison outputs by comparing the absolute value output which is not related to the attenuation output Kll among the conversion outputs and the attenuation output, respectively; and a delay n time detector comprising means for outputting a delay time of a delayed output corresponding to the absolute value conversion circuit having the absolute value conversion output closest to the attenuated output. ψ) Claim J! characterized in that the damping ratio is 0.100, which is within the range of OS! Delay time detector according to section I1. (s) The slow n time detector according to claim 1, wherein the attenuation ratio is within the range of 0.26±008. (4) The n-hour weekly detector according to any one of claims 1 to 2, wherein the input signal is a digital signal and the signal is processed by digital calculation. (6) A plurality of bands that pass only the signals of Claims Ii1 to 3, characterized in that the signal after high-frequency components are removed from the acoustic input signal by a low-pass filter is applied to the input terminal. The delay time detector according to claim 11, wherein the delay time detector is applied to an input terminal after passing through a circuit consisting of a pass/pass filter. (7) An input terminal to which a sound-human input signal is applied, a delay means for delaying the input signal and generating a plurality of delayed outputs with different delay times, and connecting the input signals to each other or the input signal and the delayed output, respectively. Multiplication means for multiplying to generate a plurality of multiplication outputs, integrating means for integrating each of the multiplication outputs to generate a plurality of integral outputs, and converting each of the above integral outputs into absolute values to generate a plurality of absolute value outputs. Absolute value converting means for generating a damped output by attenuating the absolute value converted output related to the multiplication output of the input signals among the absolute value converted output by a predetermined damping ratio; Comparing means for generating comparative outputs by comparing the absolute value output unrelated to the attenuated output and the attenuated output; An acoustic device comprising means for outputting a delayed signal of a delayed output corresponding to an absolute value U path having an absolute value output closest to the upper attenuation output, and an addition means for adding the input signal and the delayed signal. Device. (8) The delay signal has its delay time multiplied by (28±10) n
The acoustic device according to claim 1, characterized in that the audio device is added to the input signal after the signal is input. (9) an input terminal to which an acoustic input signal is applied; a delay means for delaying the input signal to generate a plurality of delayed outputs with different delay times; multiplication means for multiplying each of the above multiplication outputs to generate a plurality of multiplication outputs, an integrating means for integrating each of the above multiplication outputs to generate a plurality of integral outputs, and squaring each of the above integral outputs to generate a plurality of squared outputs. squaring means; addition means for sequentially adding the squared outputs to generate a plurality of summed outputs; attenuating means for generating a damped output by attenuating the maximum output of the summed outputs at a predetermined damping ratio; Comparing means generates comparison outputs by comparing the addition output not related to B111 to the attenuation output and the attenuation output, respectively; A delay time detector comprising means for outputting a delay time of a delay output corresponding to an addition means having a similar addition output. 120. The selected time detector according to claim 119, wherein the attenuation ratio is within the range of 0.01±0.008. 10. The delay time detector according to claim S, wherein the ap damping ratio is within the range of o, osg e, oi. (b) an input terminal to which an acoustic input signal is applied, a delay means for delaying the input signal and generating a plurality of delayed outputs with different delay times, and connecting the input signals to each other or the input signal and the delayed output, respectively; Multiplying means for multiplying to generate a plurality of multiplication outputs, integrating means for integrating each of the multiplication outputs to generate a plurality of integral outputs, and converting each of the above integral outputs into absolute values to convert them into a plurality of absolute values. An absolute value generator that generates an output. A predetermined number (plural) of the above multiple absolute value outputs in succession
Maximum value detection means that divides into units and generates an output of the maximum value of several divisions in order to find the maximum value of each division unit; Attenuating means for attenuating the digitized output at a predetermined attenuation ratio to generate an attenuated output, and comparing means for comparing the divided several maximum value outputs by the maximum value detecting means and the attenuated output to respectively generate comparison outputs. , a delayed n time detection means for outputting a delay time corresponding to a maximum value detection means related to a comparison output having a predetermined comparison result among the comparison outputs and having a maximum value detection output closest to the attenuation output; vessel. An input terminal to which an input signal is applied, a delay means for delaying the input signal and generating a plurality of delayed outputs having different delay times, and multiplying the input signals by each other or by the input signal and the delayed output, respectively. A multiplication means that generates a plurality of multiplication outputs, an integration means that integrates each of the multiplication outputs to generate a plurality of integral outputs, and an absolute value of each of the above integral outputs to generate a plurality of absolute value outputs. Absolute value means. 56 of the absolute value output; attenuation means for attenuating the absolute value output related to the multiplication output of the input signals at a predetermined attenuation ratio to generate a damped output; 56 of the absolute value output; related to the attenuated output; Comparing means for generating comparative outputs by comparing the absolute value output and the attenuated output, and continuously dividing the comparison output into a predetermined number (plurality) of units and including each division unit in the division unit n an AND means for determining and outputting from the comparison output that all the absolute value outputs are smaller than the attenuation output; A delayed time detector comprising means for outputting a delay time corresponding to a logical product means to which a delay means having a delay time closest to a delay time corresponding to a logical product means having a result belongs.