JPH067708B2 - Automatic equalizer characteristic determination method - Google Patents

Description

TECHNICAL FIELD The present invention relates to an equalizer characteristic automatic determination method, and more particularly to an equalizer characteristic automatic determination method that substantially flattens the listening characteristic based on the acoustic characteristic of a sound field.

<Prior Art> There is an audio device that obtains acoustic characteristics of a sound field and makes the listening characteristics substantially flat over the entire area based on the obtained acoustic characteristics.

Such an audio device has (1) means for requesting automatic determination of equalizer characteristics, (2) means for generating a test signal having a substantially constant level over the entire audio signal frequency, and (3) in several bands. An equalizer section for controlling the frequency characteristic of the input audio signal based on the equalizer characteristic specified by the gain or the attenuation amount, (4) a detecting section for detecting the level of each band in the sound field, and (5) the It has a control section for determining an equalizer characteristic that makes the listening characteristic in the sound field substantially flat over the entire area based on the detected level, and the test signal generating means is provided in the latter stage of the equalizer section, and a sound is generated by an equalizer characteristic automatic determination request. The level of each band of the test signal output to the field is detected, and the listening characteristic in the sound field is flattened based on the level of each band detected. Determine the equalizer characteristic to have to set the equalizer.

That is, conventionally, a test signal whose level is constant over the entire audio frequency is output to a sound field via a line amplifier, a power amplifier, and a speaker by a request for automatic determination of equalizer characteristics. Then, assuming that the level of each band of the test signal output to the sound field is detected and the acoustic characteristic obtained by using the detected level is as shown in FIG. As shown in FIG. 6B, the equalizer characteristic is determined so that the listening characteristic becomes flat, and the equalizer characteristic is set in the equalizer section. In FIG. 3, the horizontal axis represents frequency and the vertical axis represents gain (dB).

<Problems to be Solved by the Invention> However, even if the equalizer characteristic is determined by such a conventional method, the listening characteristic does not become flat, and the characteristic becomes uneven as shown in FIG. 3 (D).

This is because the test signal is not output to the sound field through the equalizer section, and moreover, the equalizer characteristic determination process is performed only once to determine the final equalizer characteristic. It should be noted that the reason why the flatness is not achieved without the equalizer section is that the frequency level adjusting section by the simulated inductor constituting the equalizer section has a certain degree around the respective center frequencies f ₁ and f ₃ as shown in FIG. Because they overlap each other with width. That is,
Even if only the frequency levels of f ₁ and f ₃ are changed as shown by the solid line and the frequency level of f ₂ is not changed, the total frequency level becomes as shown by the dotted line, and substantially the frequency level of f ₂ is also changed. Because it seems to have changed.

From the above, it is an object of the present invention to provide an equalizer characteristic automatic determination method capable of accurately flattening the listening characteristic over the entire region according to the acoustic characteristic of the sound field.

<Means for Solving Problems> The equalizer characteristic automatic determination method of the present invention is (1) provided with a means for generating a test signal having a substantially constant level over the entire frequency range of an audio signal in the preceding stage of the equalizer section, A first step of setting a predetermined equalizer characteristic in the equalizer unit by an equalizer characteristic automatic determination request and inputting the test signal to the equalizer unit; and (2)
The second step of detecting the level of each band of the test signal output to the sound field via the equalizer section, and (3) the equalizer characteristic correction data for flattening the listening characteristic based on the detected level of each band. The third step to obtain (4)
The fourth step of obtaining a new equalizer characteristic by using the equalizer characteristic correction data and the equalizer characteristic data set in the equalizer section and setting the new equalizer characteristic in the equalizer section, and (5) repeating the processing after the second step a predetermined number of times. And finally determining the equalizer characteristic to be set in the equalizer section.

<Operation> Means for requesting automatic determination of equalizer characteristics, means for generating a test signal having a substantially constant level over the entire audio signal frequency, and equalizer characteristics specified by gain or attenuation in several bands. An equalizer section for controlling the frequency characteristic of the inputted audio signal based on the above, a detecting section for detecting the level of each band in the sound field, and an equalizer for substantially flattening the listening characteristic based on the detected level. A controller for determining the characteristic is provided, and the test signal generating means is provided in the preceding stage of the equalizer section to constitute an equalizer characteristic automatic determining device.

Then, in response to the equalizer characteristic automatic determination request, for example, the equalizer characteristic in which the gain or the attenuation amount of each band is 0 dB is set in the equalizer section, and the test signal is input to the equalizer section. Then, the level of each band of the test signal output to the sound field via the equalizer unit is detected, and the equalizer characteristic for flattening the listening characteristic is determined based on the detected level of each band, and the equalizer is determined. Set the characteristics in the equalizer section. Then, the level of each band of the test signal output to the sound field is detected based on the equalizer property, and the equalizer property correction data for flattening the listening property of the sound field is obtained based on the detected level of each band. . Finally, the equalizer characteristic set in the equalizer section and the gain or attenuation of the corresponding band of the equalizer characteristic correction data are added to finally determine the equalizer characteristic set in the equalizer section.

<Embodiment> FIG. 1 is a block diagram of an equalizer characteristic automatic determining apparatus for realizing an equalizer characteristic automatic determining method according to the present invention.

In the figure, 11 is an equalizer section for controlling the frequency characteristic of an audio signal input from an input terminal IN based on predetermined equalizer characteristic data, 12 is an operating section, and a key for requesting automatic determination of the equalizer characteristic (auto Equalizer key), a key requesting frequency control based on preset equalizer characteristic data, a key requesting display of the frequency level of the output signal (spectrum analyzer display), gain or attenuation for each band is set It has various keys such as an up key and a down key provided for each band. Reference numeral 13 denotes a control unit having a microcomputer configuration, which has a processor, a RAM, a ROM, an input / output port, etc. (not shown), and performs processing in response to a request from the operation unit 12 based on a control program stored in the ROM , (1) Automatic equalizer characteristic determination process, (2) Frequency characteristic control by preset equalizer characteristic, (3) Control based on equalizer characteristic set by up key or down key, (4) Spectrum analyzer display control And so on. Reference numeral 14 denotes a pink noise generator provided in the preceding stage of the equalizer unit 11, which generates a constant level of pink noise (hereinafter referred to as a test signal) over the entire frequency band of the audio signal in the equalizer characteristic automatic determination process. Reference numeral 15 denotes a level detection unit that detects the level of each band. (1) During the equalizer characteristic automatic determination process, the level of the test signal output to the sound field (for example, in the vehicle) is detected and input to the control unit 13. Then, (2) when the spectrum analyzer is displayed, the level in each band of the audio signal output from the equalizer unit 11 is detected and input to the control unit 13. Reference numeral 16 denotes a spectrum analyzer display and a display section for selectively displaying the equalizer characteristics set in the equalizer section 11 in a bar graph, and includes LED elements (not shown) arranged in a matrix and a digit driver for driving the LED elements. , A segment driver, and the like. 17-18
Is a muting circuit, and 19 is a mixing circuit for appropriately inputting an audio signal and a test signal to the equalizer section 11. 20 is a line amplifier.

The equalizer unit 11 includes a bridge amplifier 11a and a control unit 1.
3, the equalizer characteristic (gain or attenuation of each band) is digitally set, the latch circuit 11b, n simulated inductor units 11c-1 to 11c-n provided corresponding to each band, N analog switches 11d-1 to 11d- provided corresponding to the band
It is configured to have n. The analog switches 11d-i (i = 1, 2, ..., N) provided corresponding to each band perform switching operation based on the gain (or attenuation amount) of each band set in the latch circuit 11b. The corresponding simulated inductor section 11c-n (i = 1,
2. Set the inductance value of n) to a predetermined value. The audio signal from the mixing circuit 19 is bridge amplifier 1
N simulated inductor portions 11c-, which are input to the + terminal of 1a and are set to a predetermined value by the analog switches 11d-i (i = 1, 2, ..., N).
bridge amplifier 11 via i (i = 1, 2, ..., N)
It is input to the-terminal of a. As described above, the input audio signal has its gain or attenuation controlled in each band according to the equalizer characteristic set in the latch circuit 11b, and is output to the sound field via the line amplifier 20 and a power amplifier, a speaker, etc. not shown. It That is, the graphic equalizer control (GEQ) can be executed by setting the equalizer characteristic data in the latch circuit. For example, (1) In the normal equalizer characteristic control, the equalizer control is performed by setting the gain (attenuation amount) data of each band set by the operation of the up key and down key provided for each band in the latch circuit 11b. (2) The control (PGEQ) based on the equalizer characteristic stored in the memory in advance can be performed by setting preset desired equalizer characteristic data in the latch circuit 11b.

The level detection unit 15 includes a microphone 15a that collects a sound signal (test signal) level in a sound field when the equalizer characteristic is automatically determined, and a head amplifier 15b that amplifies a microphone output.
, Bandpass filters 15c-1 to 15c-n provided for each band of n channels, and a level of the first channel, a level of the second channel, and so on in sequence based on a selection command from the control unit 13. An analog switch 15d that periodically outputs the level of the n-th channel and an AD converter 15e that converts an analog signal output from the analog switch 15d into a digital value are configured.

Now, in the spectrum analyzer display mode,
The audio signal output from the equalizer unit 11 is input to the bandpass filter 15c-i (i = 1, 2, ... N). In the spectrum analyzer display mode, the control unit 13 inputs a selection command to the analog switch 15d. As a result, the band pass filter output of each channel is periodically input to the AD converter 15e. Then, the control unit 13 reads the level (digital value) of each band AD-converted by the AD converter 15e, drives a segment driver and a digit driver (not shown) of the display unit 16 based on the read level, and the LED
The element is turned on and the level (spectrum) of each band is displayed in a bar graph.

On the other hand, in the equalizer characteristic automatic determination mode for automatically determining the equalizer characteristic for flattening the listening characteristic based on the acoustic characteristic of the sound field, the muting circuits 17, 1
8 is turned on, and a test signal is output from the pink noise generator 14, and the test signal is input to the equalizer unit 11 via the mixing circuit 19. As a result, the test signal collected by the microphone 15a is input to the bandpass filter 15c-i (i = 1, 2, ...) Through the head amplifier 15b, and the level of each band is similarly set in the control unit. 13 read.

Next, the equalizer characteristic automatic determination processing according to the present invention will be described.

(1) When it is desired to automatically determine the sound field, for example, the equalizer characteristic for flattening the listening characteristic in the passenger compartment of an automobile in which an audio device is installed, the auto equalizer key ( (Not shown). As a result, an equalizer characteristic automatic determination request signal is generated.

(2) The control unit 13 detects the pressing of the auto equalizer key and sets the equalizer characteristic data in which the gain of each band is 0 dB in the latch circuit 11b of the equalizer unit 11 (see FIG. 2 (A)). In FIG. 2, the horizontal axis represents frequency and the vertical axis represents gain (dB).

(3) Further, when the auto equalizer switch is pressed, the control section 13 inputs the muting enable signal MTES to the muting circuits 17 and 18 to turn on each muting circuit, and the pink noise generator 14
The enable signal TGS is input to the pink noise generator, and the pink noise generator generates a test signal whose level is constant over the entire frequency band of the audio signal. As a result, the input of the audio signal to the equalizer unit 11 and the input of the equalizer unit output signal to the bandpass filter 15c-i are blocked, and the test signal from the pink noise generator 14 is input to the equalizer unit 11 instead. , And the microphone output is input to the bandpass filter 15c-i.

(4) The test signal that has passed through the equalizer unit 11 is output from the output terminal OUT via the line amplifier 20, and is then supplied to the power amplifier of the audio device (not shown) to drive the speaker and output as sound in the listening room. It

(5) The test signal output to the listening room (sound field) is detected by the microphone 15a and becomes an electric signal, and the band pass filter 15 is passed through the head amplifier 15b.
c-i (i = 1, 2, 3 ...).

(6) In the equalizer characteristic automatic determination mode, the control unit 13 first sets the level of the frequency signal output from the bandpass filter corresponding to the reference channel, for example, the channel of 800 Hz, to the analog switches 15d and A.
The signal is read at a predetermined cycle via the D converter 15e, and it is checked whether the signal level has reached the set level (whether it has reached the 0 dB level). In other words, the control unit 1
3 first checks if the best tuning was done.

(7) If the best tuning is performed, the control unit inputs a selection command to the analog switch 15d, and the analog switch 15d cyclically outputs the bandpass filter output of each channel at a predetermined cycle according to the selection command.
Input to 5e.

(8) The AD converter 15e sequentially AD-converts each input analog band level, and the control unit 13 takes in the AD-converted digital value.

(9) If the control unit 13 reads the level of the entire test signal input from the microphone 15a (see FIG. 2 (B)), the listening characteristic becomes flat over the entire frequency band of the audio signal. The correction data (see FIG. 2 (C)) is determined. That is, the control unit 13 compares the output level of each bandpass filter 15c-i (i = 1, 2, ..., N) with the set level, converts the difference logarithmically, and outputs equalizer characteristic correction data for each band. Ask for. For example, for a channel having a signal level higher than the set level, the amount of attenuation largely determined by the difference is obtained, for a channel having a low signal level, a gain corresponding to the difference is obtained, and equalizer characteristic correction data composed of these data is obtained.

(10) Next, the control unit 13 sets the equalizer characteristic data set in the latch circuit 11b of the equalizer unit 11 (initially all 0 dB and stored in the built-in RAM).
And the equalizer characteristic correction data obtained in step (9) are added for each band (channel) and a new latch circuit 1 is added.
The equalizer characteristic to be set to 1b (see FIG. 2D) is obtained, stored in the built-in RAM, and set in the latch circuit. Since the equalizer characteristic initially set in the latch circuit 11b is 0 dB over the entire band (all channels), the first equalizer characteristic correction data obtained above becomes the equalizer characteristic data set in the latch circuit 11b as it is. There is.

(11) After that, the processes of steps (6) to (9) are repeated, and when the equalizer characteristic shown in FIG.
Then, the second equalizer characteristic correction data (see FIG. 2 (F)) for each channel for flattening the listening characteristic is obtained.

(12) Next, the control unit 13 adds the equalizer characteristic data set in the latch circuit 11b and the obtained second equalizer characteristic correction data for each channel, and finally adds the equalizer characteristic data to the latch circuit 11b of the equalizer 11. The equalizer characteristic to be set (see FIG. 2 (G)) is calculated and set in the latch circuit 11b.

(13) Then, the control unit 13 causes the muting circuit 17,
While turning off 18, the test signal generation of the pink noise signal generator 15 is stopped and the audio signal is restored.

From the above, the listening characteristic becomes almost flat over the entire band as shown in FIG. If a flat listening characteristic cannot be obtained by the above processing, the processing from step (6) is repeated until it becomes flat.

In the above description, the equalizer characteristic in which the gain of each band at the initial stage is 0 dB is set in the equalizer unit 11, but the present invention does not necessarily need to initialize the equalizer characteristic of all 0 dB.

<Effects of the Invention> As described above, a means for generating a test signal having a substantially constant level over the entire frequency range of the audio signal is provided in the preceding stage of the equalizer section, and a predetermined equalizer characteristic is set by the equalizer characteristic automatic determination request. A second step of inputting the test signal to the equalizer section and detecting the level of each band of the test signal output to the sound field via the equalizer section, and each detected band. The equalizer characteristic correction data for flattening the listening characteristic based on the level of
Step, fourth step of obtaining and setting an equalizer characteristic to be newly set in the equalizer section by using the equalizer characteristic correction data and equalizer characteristic data set in the equalizer section, repeating the processing after the second step a predetermined number of times The final step of determining the equalizer characteristic that is finally set in the equalizer section is configured to determine the final equalizer characteristic, so that a flat listening characteristic is ensured over the entire band according to the acoustic characteristic of the sound field. I was able to get it.

Further, according to the present invention, the equalizer characteristic for flattening the listening characteristic is stored, and when the predetermined equalizer characteristic is set from the operation panel, the stored equalizer characteristic and the set equalizer characteristic are added. By inputting the obtained characteristic to the equalizer, the listening characteristic in the sound field can be surely made the frequency characteristic intended by the user.

[Brief description of drawings]

FIG. 1 is a block diagram of an apparatus for realizing the present invention, FIG. 2 is a characteristic explanatory diagram for explaining the method of the present invention, FIG. 3 is a conventional characteristic explanatory diagram, and FIG. 4 is a conventional defect cause explanatory diagram. is there. 11 ... Equalizer, 11a ... Bridge amplifier, 1
1b ... Latch circuit, 12 ... Operating unit, 13 ... Control unit, 14 ... Pink noise generator, 15 ... Level detection unit, 15a ... Microphone, 15c-i (i = 1, 1
2, ...) ・ ・ Band pass filter, 15d ・ ・ Analog switch, 16 ・ ・ Display

Claims (1)

[Claims] 1. A means for generating a test signal having a substantially constant level over the entire frequency range of an audio signal is provided in the preceding stage of an equalizer section, and an equalizer characteristic automatic determination requesting means is provided in an operating section, whereby the equalizer characteristic automatic determination request 0d over the whole area
The equalizer characteristic of B is set in the equalizer section, the test signal is input to the equalizer section, the level of each band of the test signal output to the sound field via the equalizer section is detected, and each detected band is detected. Based on the level of, the equalizer characteristic correction data for flattening the listening characteristic is obtained, and a new equalizer characteristic is obtained using the equalizer characteristic correction data and the equalizer characteristic data set in the equalizer section and set in the equalizer section. Thereafter, the equalizer characteristic automatic determination method is characterized in that the equalizer characteristic correction data is finally determined by repeating the calculation processing of the equalizer characteristic correction data and the calculation / setting processing of a new equalizer characteristic.