JP2940585B2 - Transmission frequency characteristic correction device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a transmission frequency correction device, and more particularly to a transmission frequency characteristic correction device for obtaining a desired transmission frequency characteristic by a simple operation.

[0002]

2. Description of the Related Art In a conventional transmission frequency characteristic correction apparatus, the center frequency and the sharpness are fixed for each filter band, and when a plurality of characteristics are synthesized to obtain a desired transmission frequency characteristic, the synthesized frequency characteristic is known. I couldn't do that. In some cases, the center frequency and the sharpness can be changed for each band. In this case, the parameters of the center frequency, the sharpness and the gain are provided for each band.

[0003]

For this reason, in the former transmission frequency characteristic correction apparatus described above, since the synthesized transmission frequency characteristic cannot be known, the user will think of the desired frequency correction characteristic. There was a problem that it could not be obtained. In the latter case of the above-described conventional transmission frequency characteristic correction device, since the parameters of the center frequency, the sharpness, and the gain must be given for each band, the user must be proficient in using the frequency correction device. Thus, there is a problem that desired frequency correction characteristics cannot be obtained.

When the sharpness is determined by inputting the center frequency and the bandwidth in order to solve the above problem, a characteristic having a large gain difference over a wide band as shown in FIG. There was a problem that it was not possible.

An object of the present invention is to provide a transmission frequency characteristic correction device which can set a desired frequency correction characteristic by a simple operation.

[0006]

According to a first aspect of the present invention, there is provided a transmission frequency characteristic correcting apparatus including a plurality of filters for filtering an input signal by a transfer function whose coefficient is determined by using a center frequency, a sharpness and a gain as parameters. A variable filter section, a target characteristic data input section for inputting target characteristic data, a sharpness estimating section for inputting a bandwidth and a gain difference to estimate sharpness, and a characteristic input from the target characteristic data input section. Sending the bandwidth and gain difference to the sharpness estimating unit based on the data, and varying the filter characteristics of the variable filter unit based on the error between the filter characteristics and the characteristic data and the sharpness estimated by the sharpness estimating unit. And a control unit for setting a synthesis filter characteristic.

According to a second aspect of the present invention, there is provided a transmission frequency characteristic correcting apparatus, comprising: a variable filter section having a plurality of filters for filtering an input signal by a transfer function in which a coefficient is determined using center frequency, sharpness and gain as parameters. A target characteristic data input unit for inputting characteristic data, a sharpness estimating unit for estimating sharpness by inputting a bandwidth and a gain difference, and an optimum center frequency with a frequency having an error and a center frequency as input values. A center frequency estimating unit to be estimated, and a bandwidth and a gain difference are sent to a sharpness estimating unit based on the characteristic data input from the target characteristic data input unit. Sent to the estimation unit, and the error between the filter characteristics and the characteristic data,
A controller configured to vary the characteristics of the filter of the variable filter unit based on the sharpness estimated by the sharpness estimating unit and the center frequency estimated by the center frequency estimating unit and to set a synthesis filter characteristic. I do.

In the transmission frequency characteristic correcting device according to the first or second aspect of the present invention, a display unit for displaying the combined characteristics may be provided.

[0009]

According to the transmission frequency characteristic correction device of the present invention, target characteristic data is input from the target characteristic data input unit. Based on this input, the bandwidth and gain difference are transmitted from the control unit to the sharpness estimation unit, and the sharpness estimation unit estimates the sharpness based on the transmitted bandwidth and gain difference, and the estimated sharpness is controlled. Sent to the department
Based on the estimated sharpness, the characteristics of the filter of the variable filter unit are varied under the control of the control unit, and the synthetic filter characteristics are set. In this case, since the filter characteristics of the variable filter unit are varied according to the sharpness estimated based on the bandwidth and the gain difference, a characteristic having a large gain difference in a wide band can be approximated.

According to the transmission frequency characteristic correction device of the present invention, the target characteristic data is inputted from the target characteristic data input section. Based on this input, the bandwidth and gain difference are sent from the control unit to the sharpness estimating unit, and the frequency having an error and the center frequency are sent to the center frequency estimating unit. Based on the transmitted bandwidth and gain difference, the sharpness estimation unit estimates the sharpness. The center frequency estimating unit estimates the optimum center frequency based on the transmitted frequency having the error and the center frequency. The estimated sharpness and the optimum center frequency are sent to the control unit, and the characteristics of the filter of the variable filter unit are varied under the control of the control unit based on the estimated sharpness and the center frequency, and the combined filter characteristics are obtained. Is set. In this case, since the filter characteristic of the variable filter unit is varied according to the estimated optimum center frequency and the estimated sharpness, a characteristic having a large gain difference in a wide band can be approximated.

When the display unit is provided, the data of the synthesis filter characteristic set in the control unit is displayed on the display unit, and the use is further facilitated.

[0012]

The present invention will be described below with reference to examples. FIG.
FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

The transmission frequency characteristic correction apparatus according to the present embodiment includes a keyboard for inputting a target frequency characteristic desired by a user;
Determination of target characteristic data input unit 1 such as a light pen and an interface of an acoustic characteristic measuring device, and bandwidth data and gain difference data to be transmitted to sharpness estimating unit 3 based on data transmitted from target characteristic data input unit 1 , Center frequency estimator 4
Determination of frequency data having an error to be transmitted to the filter, determination of center frequency data, determination of data of center frequency f ₀ , sharpness factor Q, gain or attenuation of a plurality of filters, determination of coefficients of a plurality of filters, and variable filter section 6 And a control unit 2 for transferring data to the display unit 5 and data to the display unit 5.

Here, the sharpness estimating unit 3 receives the bandwidth data and the gain difference data sent from the control unit 3 and outputs the sharpness by fuzzy inference. The estimated sharpness is sent to the control unit 2. The center frequency estimating unit 4 includes the control unit 3
The frequency data having an error and the center frequency data sent from the controller are input, and the optimum center frequency is estimated. The estimated center frequency data is sent to the control unit 2.

The display unit 5 displays the data sent from the control unit 2. The variable filter unit 6 is a filter having a center frequency f ₀ , a sharpness Q, and a variable gain having a plurality of filters for performing a filtering operation using a transfer function having a coefficient determined by the control unit 3.

In the transmission frequency characteristic correction device of the present embodiment configured as described above, data based on a desired frequency characteristic input from the target characteristic data input unit 1 is controlled as filter unit setting data (hereinafter, setting data). Supply to section 2.

The control section 2 receives the setting data, sets a band number which is the number of a filter to be set, to 0, and performs a predetermined initialization operation such as turning off a limit flag. The filter setting calculation process shown in FIG.

First, all the setting data are searched, and from the setting data, the maximum value and the minimum value of the setting data, the frequency (MAX_f) which becomes the maximum value, and the frequency (MI
N_f) is determined (step S1). At this time, if there is a set value that takes the same value as the maximum value and the minimum value, the frequency (MAX_f) and the frequency (MIN_f) take the lowest frequency, respectively (step S1).

One example of the setting data is

[0020]

[Table 1] As shown in FIG.

Therefore, the maximum value is 8 (dB), the minimum value is 2 (dB), the frequency (MAX_f) is 60 (Hz),
The frequency (MIN_f) is 120 (Hz).

Next, a filter determining process is executed (step S2). The filter determination process is shown in FIG. In the filter determination process, the previously determined maximum value, minimum value, frequency (M
AX_f) and the frequency (MIN_f), the center frequency (f ₀ ), the sharpness (Q), the gain or the attenuation (GAIN), which are the parameters of the filter section coefficient, are determined, and the filter of the filter section 6 indicated by the band number is determined. The coefficients are calculated and the filter coefficients are stored in a coefficient table. Details of the filter determination processing will be described later. In the example of the setting shown in Table 1, the coefficient parameter of the band number 0 is the center frequency f
₀ is 60 Hz, sharpness Q is 1.75, GAIN is 8 dB
Get.

Next, calculation of a set band characteristic indicated by a band number and calculation of a composite characteristic are performed (step S3). FIG. 17 shows a composite characteristic diagram when band number 0 is set in the setting example in Table 1.

Next, an error calculation process is executed (step S4). In the error calculation processing, the characteristic value at the set frequency of the previously calculated composite characteristic is obtained as an approximate value. Next, band division of the synthesis characteristics is performed using the three set frequencies as one band. The following value is obtained as an error between the set value and the approximate value.

Absolute error = setting value−approximate value Conversion value = | maximum value of attenuation amount + value before conversion + 1 relative error = 1.0− (conversion value of approximation value / conversion value of setting value) The maximum value of the amount of attenuation is the maximum value of the amount of attenuation that can be set in the filter unit 6, and is -12 dB here.

At each set frequency, the set value is -6 dB
It is determined whether it is within .about. + 6 dB, and if it is within and the relative error is within 0.08, the approximation of the point is terminated. If the set value is other than -6 dB to +6 dB and the relative error is within 0.15, the approximation of the point is ended.

Further, the following value is determined by obtaining a band in which the sum of relative errors is maximum. Maximum value = maximum value of absolute error, minimum value = minimum value of absolute error, MAX_f = set frequency at which relative error is maximum, MIN_f = set frequency at which relative error is minimum, and details of error calculation processing will be described later.

The example shown in Table 1 is as follows. Maximum value = 8
dB, minimum value = 8 dB, MAX_f = 8 kHz, MIN
_F = 8 kHz.

Next, the band number is incremented (step S4-a). Subsequently, it is determined whether the approximation of all the set frequency points has been completed (step S5). If it is determined that the approximation has been completed, the coefficient of the variable filter unit 6 is changed based on the coefficient in the coefficient table (step S).
6), the filter determination processing ends.

If it is determined in step S5 that the approximation has not been completed, the filter determining process is executed again (step S7). In the example of Table 1, the coefficient parameter of the band number 1 from the maximum value, the minimum value, the frequency MAX_f, and the frequency MIN_f obtained above is the center frequency f ₀ = 8 kHz.
z, the sharpness Q = 0.5, and the gain = 8 dB.

Next, the characteristic calculation and the composite characteristic calculation of the set band indicated by the band number are performed (step S8). In the example of Table 1, the characteristics of band number 1 are shown in FIG.
FIG. 19 shows the composite characteristics of the band No. 1 and the band No. 1. Step S
Following 8, the movement number corresponding to the band number of the filter to move the center frequency f ₀ to Haibaryu (step S9).

Subsequently, an error calculation process is executed (step S10). In the example of Table 1, the maximum value is 6 dB and the minimum value is 0 d
B, frequency MAX_f = 250 Hz, frequency MIN_f
= 60 Hz. The movement number is 0, and the sum of the absolute errors is 16. Next, it is determined whether the movement number is high value (step S11). When the movement number is high value,
Jump to step S23.

If the moving number is not high value, it is determined whether or not the absolute value of the maximum absolute error is larger than the minimum value of the absolute error at the moving number (step S1).
2). If it is determined that the frequency is large, the frequency having the error is set as the set frequency at which the error has a maximum value (step S13).
If it is determined that the frequency is small, the frequency having the error is set as the set frequency at which the error has a minimum value (step S14).

Next, the sum of the current center frequency f ₀ and the absolute error is saved (step S15). Next, the frequency having the error and the center frequency f ₀ are sent to the center frequency estimating unit 4 (step S16). The center frequency estimating unit 4 estimates a new center frequency f ₀ from the set value, the approximate value, and the relative error at the frequency having the error, and sends it to the control unit 2. Details of the processing will be described later.

In the example of Table 1, the frequency having an error = 250H
z, the center frequency f ₀ = 120 Hz estimated from the center frequency f ₀ = 60 Hz is obtained. In the filter setting calculation process, a new center frequency f ₀ is received from the center frequency estimating unit 4 (step S17). The filter coefficient is calculated from the new center frequency f ₀ , the sharpness Q at the movement number, and the gain (step S18). Next, the characteristic calculation and the composite characteristic calculation of the moving band indicated by the moving number are performed (step S1).
9). FIG. 20 shows the characteristics of the moving band in the example of Table 1.
FIG. 21 shows the combined characteristics.

Next, the sum of the absolute errors is determined (step S
20). In the example of Table 1, the sum of the absolute errors in the combined characteristics after the movement of the center frequency f ₀ is 25. Subsequently, it is determined whether or not the sum of the absolute errors after moving the center frequency is smaller than the sum of the absolute errors before moving the center frequency (step S21). If it is determined to be smaller, the coefficient table of the center frequency divided by the band indicated by the movement number is changed based on the calculated coefficient (step S22). If it is determined to be larger, the coefficient is calculated again from the sharpness Q and the gain at the moving number at the center frequency f ₀ saved earlier, and the characteristics of the moving band and the combined characteristics are calculated (step S22-).
a).

In the example of Table 1, since the sum of the absolute errors calculated in step 10 is 16 and the sum of the absolute errors calculated in step S20 is 25, the center frequency f ₀ before movement is 60 Hz, Sharpness Q = 1.75, gain = 8dB
Calculate the coefficient and characteristics again. FIG. 22 shows the center frequency f.
_This shows the composite characteristics when ₀ is returned to the value before the movement. Next, the band number is incremented (step S23),
Jump to step S5.

In the example of Table 1, the band number is set to 2 and the process jumps to step S5. Here, 250 Hz, 500
Hz and 1 kHz, the approximation has not been finished yet, so that the processing after step S7 is executed. By executing the filter determination processing, the maximum value obtained in step S10 = 6 dB, the minimum value = 0 dB, and the frequency MAX_f = 250
After performing the above-described processing using hz and the frequency MIN_f = 60 Hz, the parameters of the filter are set to the center frequency f _0.
= 205 Hz, sharpness Q = 0.82, and gain = 6 dB.

Next, step S8 is executed. FIG. 23 shows the set band characteristic of band number 2, and FIG. 24 shows the composite characteristic. Next, after execution of step S9, an error calculation process is executed (step S10). This completes the approximation of all the set frequencies. Since the movement number remains high value, the process jumps to step S23. After incrementing the band number in step S23, it is determined in step S5 that all set frequency points have been approximated. next,
The coefficients of the variable filter section 6 are changed based on the coefficients in the coefficient table (step S6). Thus, the filter setting calculation process ends the process.

Next, the filter determining process will be described with reference to FIGS. In the filter determination process, it is determined whether any one of the conditions that the maximum value and the minimum value are both negative is satisfied under the condition that the absolute value of the previously determined maximum value is smaller than the absolute value of the minimum value ( Step 24) If not satisfied, a peaking filter is selected, the center frequency f ₀ is set to MAX_f, and a gain (gain) is set.
On the other hand, if any of the conditions is satisfied, a notch filter is selected, the center frequency is set to MIN_f, and the attenuation is set to the minimum value (step 25).

Next, it is determined whether or not MAX_f is higher than MIN_f (step 27). If it is determined to be high, it is next determined whether or not a peaking filter has been selected (step 28). If it is determined that the peaking filter has been selected, then it is determined whether the band number is 0 (step 29). If the band number is determined to be 0,
The gain difference in the forward direction is the value of the set value at the set frequency one point lower than MAX_f from the maximum value, and the gain difference in the reverse direction is the value of the set value at the set frequency one point higher than MAX_f from the maximum value. (Step 30).

When the band number is determined to be other than 0, the gain difference in the forward direction is set to the value of the absolute error difference at the set frequency one point lower than MAX_f from the maximum value, and the gain difference in the reverse direction is set to MAX_f from the maximum value. The value of the difference of the absolute error at a frequency higher by one point is set (step 31). Next, the bandwidth is set to the value of the difference between the center frequency f ₀ and the set frequency one point lower than MAX_f (step 32).

If it is determined in step S28 that the notch filter has been selected, it is next determined whether or not the band number is 0 (step 33). If the band number is determined to be 0, the forward gain difference is set to the value of the set value difference at the set frequency one point higher than MIN_f from the minimum value, and the reverse gain difference is set one point lower than MIN_f from the minimum value. The difference between the set values in the frequency is set as a value (step 34). If the band number is determined to be 0 or more, the positive gain difference is set to the value of the absolute error difference at the set frequency one point higher than MIN_f from the minimum value, and the reverse gain difference is one point lower than MIN_f from the minimum value. The value of the difference between the absolute errors at the set frequency is set (step 35).

Next, the bandwidth is changed from the center frequency f ₀ to MIN_
The difference between the set frequencies one point higher than f is set (step 36). If it is determined that MAX_f is lower than MIN_f, then it is determined whether or not a peaking filter has been selected (step 37). When it is determined that the peaking filter has been selected, it is next determined whether or not the band number is 0 (step 38). If the band number is determined to be 0,
The gain difference in the forward direction is the value of the set value at the set frequency one point higher than MAX_f from the maximum value, and the gain difference in the reverse direction is the value of the set value at the set frequency one point lower than MAX_f from the maximum value. (Step 3
9).

When it is determined that the band number is other than 0, the gain difference in the positive direction is set to the value of the absolute error difference at the set frequency one point lower than MAX_f from the maximum value (step 4).
0). Next, the bandwidth is set to the difference between the center frequency f ₀ and the set frequency one point higher than MAX_f (step 4).
1). If it is determined that the notch filter has been selected, then it is determined whether the band number is 0 (step 42). If the band number is determined to be 0, the gain difference in the forward direction is set to the difference between the set values at the set frequency one point lower than MIN_f from the minimum value, and the gain difference in the reverse direction is set to MIN from the minimum value.
A difference value at a set frequency that is one point higher than _f is set (step 43).

If the band number is determined to be other than 0, the gain difference in the positive direction is set to the value of the absolute error difference at the set frequency one point lower than MIN_f from the minimum value, and the gain difference in the reverse direction is set to 1 from the minimum value to MIN_f. The value of the difference of the absolute error at the point-high frequency is set (step 44). Then the value of the difference between one point lower set frequency than MIN_f bandwidth from the center frequency f ₀ (step 45). The absolute values of the gain difference in the forward direction, the gain difference in the reverse direction, and the bandwidth determined in this way are set as final values (step 46).

Next, it is determined whether or not the center frequency f ₀ is a set limit value (step 47). When it is determined that the value is the set limit value, the limit flag is turned on (step 48). Next, it is determined whether or not the gain difference in the positive direction is 0 (step 49). When the gain difference is determined to be 0, the sharpness Q is set to a predetermined value (step 50). If the gain difference is determined to be other than 0, the gain difference and the bandwidth in the forward direction and the reverse direction are sent to the sharpness estimation unit 3 (step 5).
1). The sharpness estimating unit 3 estimates the value of the sharpness Q by fuzzy inference using the forward and backward gain differences and bandwidths, and sends it to the control unit 2. Details of the processing will be described later.

Next, the filter determination process receives the value of the sharpness Q (step 52). Next, a filter coefficient is calculated from the center frequency f ₀ , the sharpness Q, and the gain (step 53). Next, after the coefficients are stored in the coefficient table (step 54), the filter determination processing ends.

The progress of the processing in the example of FIG. 8 is as follows. For band number 0, the maximum value is 4, the minimum value is 2, MAX_f is 60 Hz, MIN_f is 120 Hz, and a peaking filter is selected. The value of center frequency f ₀ = 60
Hz and a gain value = 8 dB are obtained. The gain difference in the forward direction is 6 dB, the gain difference in the reverse direction is 0 dB, the limit flag is on,
As for the bandwidth, a value of the sharpness Q = 1.75 is obtained from 60 Hz.

For band number 1, the maximum value is 8, the minimum value is 8, MAX_f is 8 kHz, and MIN_f is 8 kHz.
Peaking filter selection from z, value of center frequency f ₀ =
8 kHz and a gain value = 8 dB are obtained. The gain difference in the forward direction is 0 dB, the gain difference in the reverse direction is 0 dB, the limit flag is off, and the value of the sharpness Q = 0.5 is obtained from the bandwidth of 4 kHz.

For band number 2, the maximum value is 4d
B, minimum value is 0 dB, MAX_f is 250 kHz, MI
For N_f, the peaking filter is selected from 60 kHz, the value of the center frequency f ₀ = 250 Hz, and the value of the gain = 6 dB. The gain difference in the forward direction is 5 dB and the gain difference in the reverse direction is 2 dB
B, the limit flag is off, and the value of the sharpness Q = 0.82 is obtained from the bandwidth of 130 kHz.

Next, an error calculation process will be described with reference to FIGS. In the error calculation processing, first, band division is performed with the set frequency 3 points as one band (step 55). Next, it is determined whether the following values have been obtained for all the bands (step 56). If it is determined that all the bands have not been obtained, the following processing is performed.

The absolute error is defined as the value of the difference between the set value and the approximate value. Here, the approximate value is a characteristic value that becomes a characteristic value at a set frequency of the composite characteristic. Find the sum of absolute errors. The set value and the approximate value are converted by the following equation. Conversion value = | Maximum value of attenuation amount | + Conversion value + 1 Here, the maximum value of the attenuation amount is the maximum value of the attenuation amount that can be set in the filter unit, and is -12 dB here.

Next, relative error = 1.0- (converted value of approximate value / converted value of set value) Sum of relative errors for each band, absolute value, maximum value of relative error, minimum value of absolute and relative error The frequency at which the relative error has the maximum value and the frequency at which the relative error has the minimum value are obtained (step S57).

In the band of 4 kHz or more, when the minimum value and the maximum value of the relative error are equal to each other, the frequency at which the error takes the maximum value and the frequency at which the error takes the minimum value are both set at the center of the band.

Next, it is determined whether the set value is within -6 dB to +6 dB (step 58). If it is determined that
Next, it is determined whether the relative error at the set frequency is within 0.08 (step 59). If it is determined that it is within 0.08, the approximation at the set frequency is terminated (step 60). If it is determined that the set value is other than -6 dB to +6 dB, then it is determined whether the relative error at the set frequency is within 0.15 (step 61). If it is determined that it is within 0.15, the approximation at the set frequency is terminated (step 62).

Next, a band in which the sum of the relative errors is maximized is determined (step 63). CHK_END is turned off (step 64). Next, it is determined whether all bands have been checked or whether CHK_END is on (step 6).
5). If neither condition is satisfied, the following processing is performed in order from the band having the largest relative error sum.

It is determined whether the maximum value of the relative error is between 0.16 and 0.39 (step 66). 0.16-0.
If it is determined that it is between 39 and 39, it is next determined whether or not there is a filter in which the center frequency f ₀ is set (step 67). If it is determined that there is a filter, the control unit 2
Is the number of the band currently being checked, and
HK_END is turned on (step 68). If it is determined that there is no filter having the center frequency f ₀ set in the band or that the relative error is not between 0.16 and 0.39, the band to be checked is set to the band with the next largest error (step 69). ).

Next, the next value is set according to the value of the band where the sum of the relative errors is the largest (step 70), and then the error calculation processing ends. The maximum value is the maximum value of the absolute error, the minimum value is the minimum value of the absolute error, MAX_f is the set frequency at which the relative error is maximum, and MIN_f is the set frequency at which the relative error is minimum.

The progress of the processing in the example of Table 1 is as follows. Characteristic band division is 60Hz, 120Hz, 250
Hz is low range, 500Hz, 1kHz, 1kHz is middle range,
4 kHz, 8 kHz and 16 kHz are defined as high frequencies. For the band number 0, the sum of the absolute error is 0.382 for the 44 low frequencies, and the maximum value of the relative error is 0.31.
6. The maximum value of the absolute error is 6 dB, the minimum value of the relative error is 0.0, the minimum value of the absolute error is 0 dB, the frequency at which the relative error is maximum is 250 Hz, and the frequency at which the relative error is minimum is 60 Hz.

The sum of the relative errors for the middle band is 0.706,
The maximum relative error is 0.235 and the maximum absolute error is 4
dB, the minimum value of the relative error is 0.235, the minimum value of the absolute error is 4 dB, and the frequency at which the relative error is the maximum is 500H.
z, the frequency at which the relative error is minimized is 500 Hz.

The sum of the relative errors in the high frequency range is 1.143,
The maximum relative error is 0.381 and the maximum absolute error is 8
dB, the minimum value of the relative error is 0.381, the minimum value of the absolute error is 8 dB, the frequency at which the relative error is the maximum is 8 kHz,
The frequency at which the relative error is minimized is 8 kHz. The maximum of the sum of the relative errors is 8 dB at the maximum value, 8 dB at the minimum value, 8 kHz for MAX_f, and 8 k for MIN_f from the high band.
Hz.

For the band number 1, the sum of the absolute errors is 1
6, the sum of the relative errors is 0.382, the maximum value of the relative error is 0.316, the maximum value of the absolute error is 6 dB, the minimum value of the relative error is 0.0, and the minimum value of the absolute error is 0 dB. Is 250 Hz, and the frequency at which the relative error is minimum is 60 Hz.

The sum of the relative errors for the middle band is 0.706,
The maximum relative error is 0.235 and the maximum absolute error is 4
dB, the minimum value of the relative error is 0.235, the minimum value of the absolute error is 4 dB, and the frequency at which the relative error is the maximum is 500H.
z, the frequency at which the relative error is minimized is 500 Hz.

For the high frequency range, the sum of the relative errors is 0.190,
The maximum value of the relative error is 0.095, and the maximum value of the absolute error is 2
dB, minimum value of relative error is 0.048, minimum value of absolute error is 1 dB, and frequency at which relative error is maximum is 16 kHz.
z and the movement number are 0.

The maximum of the sum of the relative errors is as follows: the maximum value is 6 dB, the minimum value is 0 dB, and MAX_f is 2
50 kHz and MIN_f are 60 kHz.

For the band number 2, the sum of the absolute errors is 1
0, the sum of the relative errors is 0.100 for the low band, the maximum value of the relative error is 0.052, the maximum value of the absolute error is 1 dB, the minimum value of the relative error is 0.0 dB, and the minimum value of the absolute error is 0d.
B: The frequency at which the relative error is maximum is 120 Hz, and the frequency at which the relative error is minimum is 60 Hz.

The sum of the relative errors for the middle band is 0.165,
The maximum value of the relative error is 0.059, and the maximum value of the absolute error is 1
dB, the minimum value of the relative error is 0.053, the minimum value of the absolute error is -1 dB, and the frequency at which the relative error is the maximum is 250H.
z, the frequency at which the relative error is minimized is 500 Hz.

The sum of the relative errors in the high band is 0.190,
The maximum value of the relative error is 0.095, and the maximum value of the absolute error is 1
dB, the minimum value of the relative error is 0.048, the minimum value of the absolute error is 2 dB, the frequency at which the relative error is the maximum is 4 kHz,
The frequency at which the relative error is minimized is 16 Hz.

After the band number 2 is set, the approximation end condition is satisfied at all set frequencies.

Next, the processing of the center frequency setting section 4 will be described with reference to FIGS. In the processing of the center frequency estimating unit 4, it is first determined whether or not a set value (hereinafter, set value) at a frequency having an error is within -6 dB to +6 dB (step 71). Is determined to be within the range, the relative error at the next frequency having an error (hereinafter, relative error) is 0.
It is determined whether it is smaller than 1 (step 72).

If it is determined that the value is smaller, MOVE_WIDE is set to a value of 1/3 oct of the set frequency interval (hereinafter, 1 / oct).
3 oct) (step 73). If it is determined that the relative error is smaller than 0.1, it is determined whether the relative error is larger than 0.15 (step 74). MOVE_
The WIDE is set to 1 oct (step 75). If it is determined that it is smaller, MOVE_WIDE is set to 2/3 oct (step 76).

If it is determined that the set value is other than -6 dB to +6 dB, then it is determined whether the relative error is smaller than 0.2 (step 77). If it is judged to be small, MO
VE-WIDE is set to 1/3 oct (step 7)
8). If it is determined that it is larger, it is next determined whether the relative error is greater than 0.25 (step 79). If it is determined to be larger, MOVE-WIDE is set to 1 oct (step 80). MOV is judged to be smaller than 0.25
E + WIDE is set to 2/3 oct (step 81).

Next, it is determined whether or not an approximate value (hereinafter, approximate value) at a frequency having an error is smaller than a set value (step 82). If determined to be smaller, then the center frequency f ₀ is determined whether higher than the frequency with an error (step 83). If the filter is determined to be high, then it is determined whether the filter is a peaking filter (step 8).
4). When the peaking shape is determined, the new fo
From MOVE_WIDE (step 8).
5).

If it is determined that the filter is a notch filter, the new center frequency f ₀ is set to a value obtained by adding MOVE_WIDE to the center frequency f ₀ (step 86). If the center frequency f ₀ is determined to be low and then filter it is determined whether or not the peaking filter (Step 87). If it is determined that the filter is a peaking filter, the new center frequency f ₀ is moved to the center frequency f ₀ .
_WIDE is added to the value (step 88). If it is determined that the filter is a notch filter, the new center frequency f ₀ is set to the value of the difference of MOVE_WIDE from the center frequency f ₀ (step 89).

If it is determined that the approximate value is larger than the set value, it is next determined whether or not the center frequency f ₀ is higher than the frequency having an error (step 90). If it is determined to be high, it is next determined whether the filter is a peaking filter (step 91). If the peaking filter is determined, the new center frequency f ₀ is moved to the center frequency f ₀ by MOVE_WID.
A value obtained by adding E is set (step 92). If it is determined that the filter is a notch filter, the new center frequency f ₀ is set to the value of the difference of MOVE_WIDE from the center frequency f ₀ (step 93).

If it is determined that the center frequency f ₀ is low, it is determined whether the filter is a peaking filter (step 94). If it is determined that the filter is a peaking filter, the new center frequency f ₀ is set to the value of the difference of MOVE_WIDE from the center frequency f ₀ (step 95). If it is determined that the notch filter MOVE_ the new center frequency f ₀ to the center frequency f ₀
A value obtained by adding WIDE is set (step 96). Next, a new center frequency f ₀ is sent to the control unit 2 (step 97),
The center frequency estimating unit 4 ends the processing.

In the example of Table 1, only the movement number 0 is processed, the center frequency f ₀ is 60 Hz, the frequency having an error is 250 Hz, the set value is 6 dB, the approximate value is 0 dB, and the relative error is 0.316. Become. Therefore MOVE_WIDE
Gets 1 oct. Since the peaking filter is selected as the filter, the new center frequency f ₀ = f ₀ +1
oct = 120 Hz is obtained.

Next, the processing of the sharpness estimating unit 3 will be described with reference to FIG. In the processing of the sharpness estimating unit 3, first, the value of the sharpness Q is estimated by fuzzy inference using the gain difference and the bandwidth in the positive direction as input values (step 98). Next, it is determined whether or not the limit flag of the control unit 2 is turned on (step 9).
9). If it is determined that it is not on, then it is determined whether the gain difference in the reverse direction is larger than the gain difference in the forward direction (step 100).

If the correction value is determined to be large, the correction value is log (1.
9 × | reverse gain difference−positive gain difference |) +1.1, and a value obtained by multiplying the estimated sharpness Q by the estimated value is set as the final sharpness Q (step 101). If it is determined to be smaller, the center frequency f _{0 of the} control unit band number is next.
It is determined whether or not the set value in is within -6 dB to +6 dB (step 102).

If it is determined that it is within the range, it is next determined whether or not the value of the sharpness Q is higher than 1.0 (step 10).
3). If it is determined to be high, the value multiplied by the correction value is set as the final sharpness Q (step 104). When it is determined that the limit flag of the control unit 2 is on, the limit flag of the control unit 2 is turned off (step 105). Next, the sharpness Q is sent to the control unit 2.
Is sent (step 106), and the sharpness estimating unit 3 ends the processing.

The progress of the processing in the example of Table 1 is as follows. The gain difference in the forward direction is 6 dB, the gain difference in the reverse direction is 0 dB, and the bandwidth is 60 Hz. Gives 1.75. Since the limit flag of the control unit 2 is on, the limit flag is turned off, and the value of the sharpness Q is sent to the control unit 2.

The gain difference in the positive direction with respect to band number 2 is 5
The value of the sharpness Q is 0.82 from fuzzy inference from dB, the gain difference in the reverse direction is 7 dB, and the bandwidth is 130 Hz. Further, the value of the sharpness Q is sent to the control unit 2 without correction from the gain difference in the forward direction> the gain difference in the reverse direction.

In the case of the embodiment of the present invention which operates as described above, the frequency characteristic obtained when the center frequency and the bandwidth shown in FIG. In the case of the present embodiment in which the sharpness was estimated, the frequency characteristics shown in FIG. 26 were obtained. This figure 2
As is clear from FIG. 6, a characteristic having a large gain difference in a wide band can be approximated.

FIG. 27 shows an example of the characteristic in the case of the conventional graphic equalizer calculated by setting the sharpness to 1.77. In contrast, the example of the characteristic in the case of the same set value in this embodiment is shown in FIG. As shown in the figure, there are no unnecessary peaks and dips, and it is understood that the characteristics are desired by the user.

Next, a modification of the embodiment of the present invention will be described. FIG. 29 is a block diagram showing a configuration of a modification of the embodiment of the present invention.

In this modification, the same components as those of the embodiment of the present invention are denoted by the same reference numerals, and the description thereof will be omitted. The output from the microphone 7 is sent to the acoustic characteristic measuring unit 8 for frequency analysis, correction data based on the frequency analysis is supplied to the target characteristic data input unit 1, and the target data is set in the target characteristic data input unit 1.

When the audio input signal is supplied and sent to the variable filter section 6, the variable filter section 6
An audio interface 9 for inputting an output from the controller 10 and transmitting the output to a changeover switch 10, a signal generating unit 11 for transmitting an output to the changeover switch 10, an amplifier 12 for amplifying an audio signal output via the changeover switch 10, Left and right speakers 13R and 13L for reproducing the output of the amplifier 12 are provided.

The control unit 21 controls the changeover switch 10 in addition to the function of the control unit 2 and also controls the signal generation unit 1.
1 has a function of controlling signal generation.

In the above modification, the control section 21 switches the changeover switch 10 to the output side of the signal generation section 10 and drives the signal generation section 11. By driving the signal generator 11, a reference signal such as an impulse or white noise is output from the signal generator 11, and the reference signal is amplified by the amplifier 12 and reproduced by the left and right speakers 13R and 13L.

The reproduced sounds from the speakers 13R and 13L are collected by the microphone 7, converted into electric signals, and subjected to frequency analysis in the acoustic characteristic measuring section 8. The frequency analysis data is sent to the target characteristic data input unit 1, and the target data is set based on the frequency analysis data. The set target data is sent to the control unit 21.

The changeover switch 10 is switched to the sound interface 9 side by the output from the control unit 21 having received the correction data. By this switching, the filter characteristics of the variable filter unit 6 are controlled by the same operation as that of the embodiment of the present invention, and the frequency characteristics of the audio input signal are corrected in the variable filter unit 6, and the sound having the corrected frequency characteristics The output signal is amplified by the amplifier 12 and reproduced by the speakers 13R and 13L.

As described above, the correction of the indoor frequency characteristic is automatically performed, and the number of bands of the correction filter is limited so that the frequency characteristic desired by the user can be obtained by the filters of the remaining bands. it can.

[0094]

As described above, according to the present invention, a simple operation of inputting the frequency characteristic data desired by the user to the target characteristic data input section can be performed even if the user is not familiar with the operation. A desired frequency correction characteristic can be easily obtained, and the operability is improved.

Further, even if a characteristic having a large gain difference is specified in a wide band, there is an effect that a frequency characteristic of an approximate characteristic can be obtained. Since the synthesized frequency characteristic is displayed, it is easy to determine whether the frequency characteristic desired by the user has been achieved, and there is also an effect that the operation of the user can be easily confirmed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 3 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 4 is a flowchart for explaining the operation of an embodiment of the present invention.

FIG. 5 is a flowchart for explaining the operation of an embodiment of the present invention.

FIG. 6 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 7 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 8 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 9 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 10 is a flowchart for explaining the operation of an embodiment of the present invention.

FIG. 11 is a flowchart for explaining the operation of an embodiment of the present invention.

FIG. 12 is a flowchart for explaining the operation of an embodiment of the present invention.

FIG. 13 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 14 is a flowchart for explaining the operation of the embodiment of the present invention;

FIG. 15 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 16 is a flowchart for explaining the operation of one embodiment of the present invention;

FIG. 17 is a frequency characteristic diagram for explaining the operation of the embodiment of the present invention.

FIG. 18 is a frequency characteristic diagram for explaining the operation of one embodiment of the present invention.

FIG. 19 is a frequency characteristic diagram for explaining the operation of one embodiment of the present invention.

FIG. 20 is a frequency characteristic diagram for explaining the operation of the embodiment of the present invention;

FIG. 21 is a frequency characteristic diagram for explaining the operation of one embodiment of the present invention.

FIG. 22 is a frequency characteristic diagram for describing the operation of one embodiment of the present invention.

FIG. 23 is a frequency characteristic diagram for describing the operation of the embodiment of the present invention.

FIG. 24 is a frequency characteristic diagram for explaining the operation of the embodiment of the present invention.

FIG. 25 is a frequency characteristic diagram in the case of a conventional example.

FIG. 26 is a frequency characteristic diagram for explaining the operation of the embodiment of the present invention;

FIG. 27 is a frequency characteristic diagram in the case of a conventional example.

FIG. 28 is a frequency characteristic diagram for explaining the operation of the embodiment of the present invention;

FIG. 29 is a block diagram showing a configuration of a modification of the embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Target characteristic data input part 2 and 21 control part 3 Sharpness estimation part 4 Center frequency estimation part 5 Display part 6 Variable filter part

Claims (3)

(57) [Claims] 1. A variable filter section having a plurality of filters for filtering an input signal by a transfer function in which a coefficient is determined using a center frequency, a sharpness and a gain as parameters, and a target characteristic data input section for inputting target characteristic data. A sharpness estimating unit that estimates a sharpness by inputting a bandwidth and a gain difference, and sends a bandwidth and a gain difference to the sharpness estimating unit based on the characteristic data input from the target characteristic data input unit, and A transmission unit comprising: a control unit configured to vary a filter characteristic of a variable filter unit based on an error between a filter characteristic and characteristic data and a sharpness estimated by a sharpness estimation unit to set a synthetic filter characteristic. Frequency characteristic correction device. 2. A variable filter section having a plurality of filters for filtering an input signal by a transfer function whose coefficient is determined using a center frequency, a sharpness and a gain as parameters, and a target characteristic data input section for inputting target characteristic data. And a sharpness estimator for estimating sharpness by inputting a bandwidth and a gain difference; a center frequency estimator for estimating an optimal center frequency using a frequency having an error and a center frequency as input values; and inputting target characteristic data. The bandwidth and gain difference are sent to the sharpness estimating unit based on the characteristic data input from the unit, the frequency and the center frequency having an error are sent to the central frequency estimating unit based on the characteristic data, and the filter characteristic and the characteristic data are sent. Of the variable filter unit on the basis of the difference between the sharpness estimated by the sharpness estimating unit and the center frequency A transmission frequency characteristic correction device, comprising: a control unit for setting a synthesis filter characteristic by changing a characteristic of the filter. 3. The transmission frequency characteristic correction apparatus according to claim 1, further comprising a display unit for displaying data of the set synthesis filter characteristic.