JPH09331223A - Signal processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a signal processor which enables a diversified signal processing. SOLUTION: The compressing characteristics of each compression circuit 2a to 2n is made to be variably set at every band by dividing input signals IN into a plurality of bands by using band-pass filters 1a to 1n and compressing the signals by using the circuits 2a to 2n at each divided band, and then, synthesizing and outputting the signals after the levels of the signals in each band is variably set. In addition, the circuits 2a to 2n are constituted so that the circuits 2a to 2n can be used as a digital signal processor which performs arithmetic operation, y=a 1-1(1-[x])<n> } on input signals (x). The [x], n, and (a) respectively represent the absolute value of the input signal (x), the order specifying the compressing characteristics, and a coefficient which is '1' when x>=0 and '-1' when x<0.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a signal processing device capable of easily and variously changing an input signal, particularly an audio signal.

[0002]

2. Related Background Art A graphic equalizer is used to correct the frequency characteristic in an audio signal reproduction environment where the frequency characteristic is not flat. This type of graphic equalizer is generally configured to divide an input audio signal into a plurality of bands, adjust the signal level for each frequency band, and then synthesize and output them. In addition, so-called sound coloring is also performed by conversely utilizing this property to increase or decrease the signal level in a specific band.

On the other hand, in a car audio device, it is difficult to hear a small level of reproduced sound due to an engine sound, a wind noise, etc., so that the output signal level is varied according to the input signal level thereof. This processing gives a large gain to a small level signal and a small gain to a large level signal to give the input / output characteristics a compression characteristic, and is called compression.

Incidentally, this compressing process is performed in an analog manner by utilizing the square characteristic of a semiconductor, or digitally by using a digital signal processor (DSP). However, the above-mentioned analog method has an advantage that the compression amount can be continuously changed, but in general, the sound quality characteristics are significantly deteriorated, and the compression characteristics depend on the semiconductor element characteristics. In addition, in the digital system,
Although the arithmetic processing is generally very complicated, there is basically an advantage that the frequency characteristic (sound quality) of the reproduced signal does not deteriorate.

[0005]

The gain adjustment processing for each band by the graphic equalizer and the compression processing (compress) by the compressor described above are very important in changing the output characteristics of the audio signal and making a so-called sound. Play a big role. However, these signal processing functions are independent of each other, and in general, it is rarely considered to use them together.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a signal processing device which enables simple and effective signal processing with greater variety. is there.

[0007]

In order to achieve the above-mentioned object, a signal processing apparatus according to the present invention divides an input signal into a plurality of bands by using a bandpass filter, and for each of the divided signals in each band. After performing compression processing using the compression circuit and variably setting the level of the compressed signal in each band, it is assumed that they are combined and output. The feature is that it can be variably set for each.

That is, not only the levels of the signals in the plurality of divided frequency bands are variably set, but also the compression characteristics to be given to the signals of the respective bands are variably set, so that the reproduced signal having a variety of adjustments It is characterized by having been obtained.

According to a second aspect of the present invention, in particular, a compression circuit used for the compression processing in each of the above-mentioned bands is provided for the input signal x with y = a {1-1 (1- [x]). It is characterized in that it is realized by a digital signal processor that obtains an output signal y by performing a calculation of ⁿ }. Where [x] is the absolute value of the signal x, n is the order that defines the compression characteristics, and a is a coefficient that takes a value <1> when x ≧ 0 and a value <−1> when x <0. is there.

[0010]

DETAILED DESCRIPTION OF THE INVENTION A signal processing apparatus according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of a signal processing device realized as a graphic equalizer having a compress function,
Reference numerals 1a, 1b, to 1n are bandpass filters (BPF) that divide the input signal IN into a plurality of frequency bands. Specifically, these bandpass filters 1a, 1b,
.About.1n equally divides the input audio signal having a frequency band of 20 Hz to 20 kHz into 5 or 9 frequency bands. The number of frequency bands to be divided may be set according to the specifications.

Each of the band pass filters 1a, 1b, ... 1
n digital signal processor (D
SP) 2a, 2b, to 2n are configured, for example, as shown in FIG. 2, and compress the signals of the divided frequency bands under individually set compression characteristics. It is what you do. This compressing process will be described later. Then, the variable gain amplifier (VGA) 3a,
3b, 3n are for individually variably setting the gains for the signals subjected to the compression processing in the respective bands.
Each signal subjected to these gain adjustments is a combiner (ADD).
4 is combined and output as the signal OUT of the original frequency band.

As shown in FIG. 2, the digital signal processor 2 includes a positive / negative determination section 11, a coefficient processing section 12,
Absolute value processing unit 13, first subtraction unit 14, exponentiation processing unit 1
5, a second subtraction unit 16 and a multiplication processing unit 17 are provided. Then, y = a {1-1 (1- [x]) ⁿ } is applied to the input signal x to obtain the compressed output signal y. In the above equation, the absolute value symbol is shown as [] for convenience. Therefore, [x] indicates the absolute value of the signal x. Further, n is an order defining the compression characteristic, and is variably set according to the compression characteristic to be set. Further, a is a coefficient set according to the polarity of the input signal x, and <1> when x ≧ 0 and <− when x <0.
1> is set.

Specifically, the positive / negative determination unit 11 determines whether the positive / negative of the input signal x in the divided frequency band is, for example, <x ≧ 0>, and the coefficient processing unit 12 The value of the coefficient a is set according to the determination result. That is, as described above, when the input signal x is <x ≧ 0>, the coefficient conversion processing unit 12 sets the value of the coefficient a to <1>.
When <x <0>, the value of the coefficient a is set as <-1>. The value of this coefficient a is given to the multiplication processing unit 17 as polarity information for the output signal y.

On the other hand, the absolute value processing unit 13 obtains the absolute value of the amplitude (level) of the input signal x as [x].
The first subtraction processing unit 14 calculates the absolute value [x] by a constant <1>.
By subtracting the maximum amplitude value of the input signal x from <1>
The complement is calculated when normalized as. In accordance with this complement <1- [x]>, the exponentiation processing unit 15 sets the order set in order to logarithmically give the attenuation amount according to the level of the input signal x and set the compression characteristic according to the signal level. The above complement <1- [x]> is exponentiated according to n. Specifically, when the order n is <2>, the exponentiation process of <1- [x]> × <1- [x]> = <1- [x]> ² is executed, and the order n is When <3>, <1- [x]> x <1- [x]> x <1- [x]> = <1-
[x]> ³ exponentiation processing is performed, and when the order n is <4>, <1- [x]> x <1- [x]> x <1- [x]> x <1 −
The exponentiation process is executed with [x]> = <1- [x]> ⁴ . That is, the amount of attenuation to be given to the input signal x is obtained according to the signal level by this exponentiation process. The inverse compression characteristic is set such that the higher the signal level, the smaller the attenuation amount, and the lower the signal level, the larger the attenuation amount.

The second subtraction processing unit 16 subtracts the attenuation amount obtained as described above from the normalized value <1> to obtain the attenuation amount to be actually given to the input signal x. It is something to set. By this subtraction processing, the compression characteristic as shown in FIG. 3 is set such that the higher the level of the input signal x is, the larger the attenuation amount is, and the lower the signal level is, the smaller the attenuation amount is, and according to the level of the input signal x. The variable signal is obtained as 1- <1- [x]> ⁿ . The signal output from the second subtraction processing unit 16 is guided to the multiplication processing unit 17, is multiplied by the above-mentioned coefficient a, is converted into the output signal y having the polarity corresponding to the polarity of the input signal x, and is output. .

The DS thus carrying out the above-mentioned signal processing
According to P2, the input signal x can be compressed under a compression characteristic defined by the order n by a simple process, that is, without a division process. That is, the input signal x can be compressed according to the signal level.

Therefore, according to the present apparatus, the signal divided into a plurality of frequency bands is subjected to a compression process for each band, the gain for the signal is variably set, and then the signals are combined and output. Since the output characteristics of the signal component in each of the above frequency bands can be set individually, it is possible to variably set the compression characteristics and then variably set the overall gain. Therefore, for example, it is possible to easily perform a special signal correction such as increasing the change in the signal level of another frequency band while making the signal level of a specific frequency band substantially constant. In other words, while utilizing the function as a graphic equalizer, it is possible to change the compression characteristics in each band individually and obtain an output signal that is suitable for the audio signal playback environment or intentionally colored (tone change). Becomes

The present invention is not limited to the above embodiment. For example, all the functions shown in FIG.
It is also possible to realize it with P, and conversely D shown in FIG.
It is also possible to set the compression characteristic of each frequency band using a log amplifier (logarithmic amplifier; Log-Amp.) Instead of SP. When actually configuring the device, the DSP 2 or V
Of course, it is also possible to appropriately provide a bypass circuit for the GA3. Further, it is possible to variably set the gain for each signal in the divided frequency band and then subject the compressed processing to this. In short, the present invention can be variously modified and implemented without departing from the gist thereof.

[0019]

As described above, according to the present invention, the compression processing is performed for each signal divided into a plurality of bands, and the gain is variably set, and then the signals of the respective bands are combined and output. Since it is possible to give different compression characteristics to each band while fully utilizing the function as a graphic equalizer, it is possible to give a signal change that could not be expected until now, and to reproduce a variety of signals. Is possible. In particular, the audio signal can be given special coloring, and the configuration of the device itself is simple, and the signal processing effect is large.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a signal processing device according to an embodiment of the present invention.

2 is a diagram showing a configuration example of a compress circuit by a digital signal processor used in the signal processing device shown in FIG.

FIG. 3 is a diagram showing an example of compression characteristics of the compress circuit shown in FIG.

[Explanation of symbols]

1a, 1b, to 1n band pass filter (BPF) 2a, 2b, to 2n Digital signal processor (DS
P; Compress circuit 3a, 3b, to 3n Variable gain amplifier (VGA) 4 Combiner 11 Positive / negative determination unit 12 Coefficient conversion processing unit 13 Absolute value processing unit 14 First subtraction unit 15 Exponentiation processing unit 16 Second subtraction Part 17 Multiplication processing part

Claims (2)

[Claims] 1. A bandpass filter group that divides an input signal into a plurality of bands, and a compress circuit group that variably sets input / output characteristics for signals of the respective divided bands and compresses each of the signals. , After variably setting the level of the compressed signal of each band,
A signal processing device comprising: an output circuit for combining and outputting. 2. The compression circuit is realized by a digital signal processor, and y = a {1-1 (1- [x]) ⁿ } for an input signal x, where [x] is The absolute value of the signal x, n is the order that defines the compression characteristic, and a is an output signal y that is a coefficient that takes a value <1> when x ≧ 0 and a value <−1> when x <0. The signal processing device according to claim 1, wherein the signal processing device is obtained.