JPH10172240A - Audio signal reproducing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a variable control of the frequency characteristic of a compression-encoded audio data to a plurality of desired frequency characteristics, and at the same time reduce the amount of calculation. SOLUTION: A compression-encoded audio data supplied from external circuit via an input terminal 1 is converted to the audio data in the frequency range through inverse quantization process by an inverse quantizing section 3 and this audio data is then supplied to a filter section 4. To the filter section 4, the frequency characteristic data of the desired frequency characteristic is being supplied to a frequency characteristic setting section 5 and the filter section 4 samples this frequency characteristic data in the frequency resolution of the audio data in the frequency region. Each sample is normalized with 0dB as '1' (reference) to form each filtering coefficient of above frequency resolution. Each filtering coefficient is multiplied to each sample of the audio data in the frequency region to execute the filtering process to each audio data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to, for example, MPEG (Moving Picture Experts Group) and Dolby AC-
The present invention relates to an audio reproducing apparatus that reproduces audio information that has been subjected to high-efficiency compression encoding by three methods or the like, and more particularly to an audio reproducing apparatus that can change the frequency characteristics of audio information to be reproduced and output to a plurality of desired frequency characteristics.

[0002]

2. Description of the Related Art At present, analog audio signals are digitized and the MPEG system (Moving Picture Exposure) is used.
erts Group), Dolby AC-3 system, etc., encoding technology for transmitting audio data that has been compressed and encoded with high efficiency or recording it on a recording medium such as an optical disk, and its decoding technology have been established and used in various fields. Have been.

More specifically, when encoding an analog audio signal, the analog audio signal is sampled by a sampling clock of a predetermined frequency to form audio data as information in a time domain. The data is converted into audio data (DCT coefficients) in the frequency domain using a cosine transform method (MDCT), a sub-band method, or the like. The audio data in this frequency domain has a certain frequency resolution, and the mantissa part is M and the exponent part is E.
× 2 ＾ E (M × expE) ”.

Then, the voice data in the frequency domain is
Quantization is performed using a quantization coefficient corresponding to the audio data, bit allocation is performed using a so-called masking effect based on human auditory characteristics, and then converted into a predetermined format. As a result, audio data obtained by compression-coding the analog audio signal can be formed.

[0005] Next, an audio reproducing apparatus for decoding the audio data compressed and encoded as described above is configured as shown in FIG. 7, and receives the bit stream of the audio data compressed and encoded as described above. The power is supplied to the unpack unit 101 via the terminal 100.

[0006] The unpacking unit 101 divides the bit stream of the audio data based on the bit allocation information added to the audio data, and supplies this to the inverse quantization unit 103.

[0007] The inverse quantization unit 103 calculates the quantization coefficient based on the quantization coefficient used when the audio data is compression-coded.
The audio data from the unpacking unit 101 is subjected to an inverse quantization process to form audio data in the frequency domain. Then, the audio data in this frequency domain is supplied to the inverse conversion unit 103.

[0008] The inverse transform unit 103 reproduces time domain audio data by performing inverse MDCT processing or the like on the frequency domain audio data reproduced by the inverse quantization processing, and supplies this to the digital filter 104.

The digital filter 104 is composed of a high-speed arithmetic processing circuit such as a digital signal processor (DSP). The digital filter 104 multiplies the samples of the audio data in the time domain by several taps of the digital filter. An arithmetic operation is performed to convert the audio data in the time domain into a set frequency characteristic, and this is output via the output terminal 105.

The audio data output via the output terminal 105 is amplified by an amplifier circuit at a predetermined gain, converted into an analog signal by a digital / analog conversion circuit (D / A conversion circuit), and supplied to the speaker device. Thus, an acoustic output corresponding to the reproduced audio data can be obtained via the speaker device.

[0011]

Here, in the reproduction of the compression-encoded audio data, restrictions on devices for reproduction, such as a dynamic range, restrictions on a reproduction environment, reproduction conditions, and a person who listens to reproduced audio For various reasons, such as the preference of the user, there is a demand to change the frequency characteristics of the audio signal to be reproduced. For example, when a speaker device that cannot input a high-frequency audio signal is used, it is necessary to remove the high-frequency component of the audio signal and supply it to the speaker device.

However, since the conventional audio reproducing apparatus performs variable frequency characteristic processing based on a filtering characteristic preset in the digital filter 104, audio data having a plurality of desired frequency characteristics is obtained. I couldn't do that.

In the conventional audio reproducing apparatus, when obtaining audio data having a plurality of desired frequency characteristics, an IIR filter (InR) is required for each frequency characteristic.
finiteImpulse Response) or FIR filter (Finite
Impulse Response) is required and it is not realistic.

When changing the frequency characteristics of audio data, the conventional audio reproducing apparatus needs to perform a product-sum operation for several taps of the digital filter 104 on each sample of audio data in the time domain. Therefore, the amount of calculation increases, and when the digital filter 104 is configured by a DSP, there is a problem that the load on the DSP increases.

When there are a plurality of desired frequency characteristics, filter coefficients of a digital filter are required for each frequency characteristic, and a large number of memories (or a large capacity) are required to store these filter coefficients. Memory).

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has been made to reduce the amount of calculation and the memory required for storing filter coefficients without increasing the compression-encoded audio data into a plurality of desired frequency characteristics. It is an object of the present invention to provide a sound reproducing device capable of changing and reproducing.

[0017]

In order to solve the above-mentioned problems, a sound reproducing apparatus according to the present invention is provided with a frequency characteristic setting means for setting a desired frequency characteristic and a frequency characteristic setting means for setting the desired frequency characteristic. Frequency characteristic variable means for variably controlling the frequency characteristic of audio information in an encoded frequency domain supplied from the outside in accordance with the frequency characteristic, and the frequency characteristic variably controlled by the frequency characteristic variable means. And a time-domain voice forming means for converting the voice information into time-domain voice information and outputting the converted voice information.

Such an audio reproducing apparatus variably controls the encoded audio information supplied from the outside to the frequency characteristic set by the frequency characteristic setting means at the stage of the audio information in the frequency domain. Then, the audio information in the frequency domain in which the frequency characteristics are variably controlled is converted into audio information in the time domain by the time domain audio forming means and output.

Thus, the coded audio information supplied from the outside can be freely variably controlled to a plurality of desired frequency characteristics, and the frequency characteristic variable processing can be performed on the time-domain audio information. It is possible to variably control the frequency characteristics more easily than by applying.

Therefore, a memory for storing each filter coefficient for variably controlling the frequency characteristic can be eliminated. Further, when the frequency characteristic changing means is realized by a high-speed operation processing circuit such as a digital signal processor (DSP), the amount of operation performed by the DSP can be reduced.
The load on the SP can be significantly reduced.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, preferred embodiments of an audio reproducing apparatus according to the present invention will be described in detail with reference to the drawings.

First, an audio reproducing apparatus according to a first embodiment of the present invention converts a bit stream of compression-encoded audio data supplied through an input terminal 1 as shown in FIG. An unpacking unit 2 that divides audio data (MDCT coefficients) in the frequency domain by performing an inverse quantization process based on quantization coefficients of audio data divided by the unpacking unit 2 based on bit allocation information attached to the data. And an inverse quantization unit 3 for reproducing.

Further, the audio reproducing apparatus includes a frequency characteristic setting unit 5 for setting a desired frequency characteristic, and a filtering process for audio data in the frequency domain based on the frequency characteristic set by the frequency characteristic setting unit 5. And a filter unit 4 that performs frequency characteristic change by applying the inverse MD to the audio data in the frequency domain whose frequency characteristic is changed by the filter unit 4.
An inverse conversion unit 6 converts the audio data in the frequency domain into the audio data in the time domain by performing the CT processing, and outputs the converted audio data via the output terminal 7.

As the compression-encoded audio data supplied via the input terminal 1, for example, analog audio signals are digitized by sampling with a sampling clock of a predetermined frequency to form audio data in the time domain. The voice data in the time domain is subjected to a modified discrete cosine transform process (MDCT process) to form voice data in the frequency domain, and the voice data in the frequency domain is quantized using an optimal quantization coefficient, and is varied. The audio data formed by performing bit allocation by the long encoding process is supplied.

It should be noted that this audio data is transmitted or recorded with the quantization coefficient selected at the time of the quantization processing, the bit allocation information allocated at the time of the variable length coding processing, and the like. Has become.

Next, the operation of the audio reproducing apparatus according to the first embodiment having such a configuration will be described.

First, when the audio data compressed and encoded as described above is supplied to the unpacking unit 2 via the input terminal 1, the unpacking unit 2 performs the processing based on the bit allocation information added to the audio data. To perform variable length decoding processing,
The bit stream is divided and supplied to the inverse quantization unit 3.

The inverse quantization unit 3 performs an inverse quantization process on the audio data based on the quantization coefficient used when quantizing the audio data, so that a constant value as shown in FIG. The audio data (MDCT coefficient) in the frequency domain composed of the samples decomposed at the frequency resolution is reproduced and supplied to the filter unit 4.

Here, a plurality of frequency characteristic data for setting a desired frequency characteristic is stored in the frequency characteristic setting section 5 in advance, and the frequency characteristic setting section 5 is selected by the user from among them. The frequency characteristic data is supplied to the filter unit 4.

For example, from the frequency characteristic data,
As shown in FIG. 2 (b), it rises gently in the low range,
When a frequency characteristic that shows stability in the middle band and sharply falls in the high band is selected (when a frequency characteristic that emphasizes stability in the middle band is selected), the frequency characteristic data indicating this is sent to the filter unit 4. Supplied.

The audio reproducing apparatus can set a desired frequency characteristic in the form of an external input. The user can set a desired frequency characteristic as a desired low-pass filter (LPF) as shown in FIG. Is set, the cut-off frequency f1 in the high frequency range is input.

When the cut-off frequency f1 is input, the frequency characteristic setting section 5 sets a stop frequency f2 higher than the cut-off frequency f1. Then, at a frequency lower than the cutoff frequency f1, the frequency characteristic is 0 dB, and from the cutoff frequency f1 to the rejection frequency f2, the frequency characteristic decreases linearly to -∞dB, and is higher than the rejection frequency f2. The frequency characteristic data of the frequency characteristic in which the frequency is fixed at −∞ dB is formed and supplied to the filter unit 4.

The frequency characteristics of the frequency lower than the cutoff frequency f1 and the frequency characteristics of the frequency higher than the stop frequency f2 are “0 dB” or “−Δd”, respectively.
Instead of a constant level at "B", an arbitrary preset level may be used. Similarly, the cutoff frequency f
The frequency characteristic between 1 and the stop frequency f2 may be an arbitrary frequency characteristic set in advance, such as a curve or stepwise decrease, instead of a linear decrease.

Next, when setting a high-pass filter (HPF) as shown in FIG. 4 as a desired frequency characteristic, the user inputs a low-frequency cutoff frequency f4.

When the cutoff frequency f4 is input, the frequency characteristic setting section 5 sets the stop frequency f3 to a frequency lower than the cutoff frequency f4. And
0 dB at a frequency higher than the cutoff frequency f4
With the constant frequency characteristic of the above, the frequency characteristic increases linearly from the stop frequency f3 to the cutoff frequency f4 to 0 dB, and the frequency characteristic data of the frequency characteristic that becomes constant at −∞ dB at frequencies lower than the stop frequency f3 is formed. This is supplied to the filter unit 4.

The frequency characteristics higher than the cut-off frequency f4 and the frequency characteristics lower than the stop frequency f3 are "0 dB" or "-Δd", respectively.
Instead of a constant level at "B", an arbitrary preset level may be used. Similarly, the frequency characteristic between the stop frequency f3 and the cutoff frequency f4 is not linearly increased, but is set as an arbitrary frequency characteristic set in advance such as, for example, a curved or stepwise increase. Is also good.

Next, when the user sets a band-pass filter (BPF) as shown in FIG. 5 as a desired frequency characteristic, the low-frequency cutoff frequency f6 and the low-frequency cutoff frequency f6, which indicate the lower limit of the lower frequency range and the upper limit of the higher frequency range, respectively. Wide-range cutoff frequency f7
Input.

When the low frequency cutoff frequency f6 and the wide frequency cutoff frequency f7 are input, the frequency characteristic setting unit 5 sets a low frequency rejection frequency f5 lower than the low frequency cutoff frequency f6 and sets a high frequency. A wide-range rejection frequency f8 higher than the frequency cut-off frequency f7 is set.

At a frequency higher than the cutoff frequency f6 and lower than the cutoff frequency f7, 0 dB
With the constant frequency characteristic of, the frequency increases linearly from the stop frequency f5 to the cutoff frequency f6, and this stop frequency f5
At a frequency lower than this, the frequency characteristic becomes -∞dB, and decreases linearly from the cutoff frequency f7 to the stop frequency f8, and at a frequency higher than the stop frequency f8, -∞dB.
The frequency characteristic data having the frequency characteristic of dB is formed and supplied to the filter unit 4.

The frequency characteristics of the frequency between the cutoff frequency f6 and the cutoff frequency f7 are as follows:
Instead of a fixed level of “0 dB”, an arbitrary level set in advance may be used. Similarly, the stop frequency f
The frequency characteristics of the frequency lower than 3 and the frequency characteristics of the frequency higher than the stop frequency f8 may not be set to a constant level of “−∞ dB”, but may be set to any predetermined level. Further, the frequency characteristics between the cutoff frequency f5 and the cutoff frequency f6 and the frequency characteristics between the cutoff frequency f7 and the cutoff frequency f8 are not linearly reduced, but are, for example, curved or stepwise. It may be an arbitrary frequency characteristic set in advance, such as decreasing the frequency characteristic.

Next, when the user sets a band rejection filter as shown in FIG. 6 as a desired frequency characteristic,
A first cutoff frequency f9 and a second cutoff frequency f12 indicating a band to be cut off are input.

When the first and second cutoff frequencies f9 and f12 are input, the frequency characteristic setting section 5 sets the first stop frequency f10 to a frequency higher than the first cutoff frequency f9. Then, the second stop frequency f11 is set to a frequency higher than the first stop frequency f10 and lower than the second cutoff frequency f12.

Then, the first cutoff frequency f9
At a lower frequency and a frequency higher than the second cutoff frequency f12, the first frequency characteristic is 0 dB and the first frequency characteristic is 0 dB.
From the cutoff frequency f9 to the first cutoff frequency f10 linearly decreases to −∞ dB, and from the second cutoff frequency f11 to the first cutoff frequency f12 linearly increases to 0 dB. In the frequency characteristic, -∞d between the first stop frequency f10 and the second stop frequency f11.
The frequency characteristic data which becomes the constant frequency characteristic of B is formed and supplied to the filter unit 4.

The frequency characteristics of the first cut-off frequency f9 or lower and the second cut-off frequency f
The frequency characteristics of the twelve or more frequencies may not be a fixed level at “0 dB”, but may be any preset level. Similarly, from the first stop frequency f10 to the second
The frequency characteristic up to the stop frequency may not be a constant level of “−「 dB ”but may be an arbitrary level set in advance. Further, the first cutoff frequency f
9 and the first cutoff frequency f10, and the frequency characteristic between the second cutoff frequency f11 and the second cutoff frequency f12 decrease linearly,
Instead of increasing the frequency characteristic, an arbitrary frequency characteristic may be set in advance such as decreasing or increasing in a curved or stepwise manner.

Next, when the desired frequency characteristic is set by the user in this way, the filter unit 4
As shown by the black dots in FIG. 3C, the frequency characteristic data is sampled in accordance with the frequency resolution to form discrete frequency characteristic data. For example, 0 dB frequency characteristic data is set to “1” (reference), and each frequency characteristic data is set as a reference. By performing a normalization process on the frequency characteristic data for each resolution, a filtering coefficient for each frequency resolution is formed. Then, the filtering coefficient is multiplied by the audio data having the corresponding frequency resolution.

As a result, the frequency characteristic of the audio data in the frequency domain as shown in FIG.
As shown in (2), it is possible to variably control the frequency characteristic set by the user, which is a frequency characteristic in which the middle region is kept as it is, the low region rises gently, and the high region falls sharply. The audio data in the frequency domain whose frequency characteristic is variably controlled is supplied to the inverse converter 6.

The inverse transform unit 6 performs inverse MDCT processing on the filtered audio data in the frequency domain, thereby converting the audio data in the frequency domain into the audio data in the time domain. Output.

The audio data in the time domain output via the output terminal 7 is converted into a digital / analog conversion circuit (D /
(A conversion circuit), and is supplied to the speaker device as an audio signal. Thereby, an acoustic output corresponding to the frequency characteristic set by the frequency characteristic setting unit 5 can be obtained through the speaker device.

As is clear from the above description, the audio reproducing apparatus according to the first embodiment of the present invention converts the compression-encoded audio data into the desired frequency characteristic set by the frequency characteristic setting section 5. Can be variably reproduced.
For this reason, it is not necessary to provide a plurality of digital filters for realizing each desired frequency characteristic, and the cost of the audio reproducing apparatus can be reduced by reducing the number of components and simplifying the configuration.

Further, instead of performing filtering processing on the audio data in the time domain reproduced by the inverse transform unit 6, the audio data in the frequency domain reproduced by the inverse quantization unit 3 preceding the inverse transform unit 6 is processed. Each MDC that is a sample of data
Since the frequency coefficient is variably controlled by multiplying the T coefficient by a filtering coefficient for each frequency resolution, it is simpler than variably controlling the frequency characteristic of the time-domain audio data at the subsequent stage of the inverse transform unit 6. The frequency characteristics can be variably controlled.

When a desired frequency characteristic is input by the user, for example, in the case of a low-pass filter, the frequency between the cutoff frequency f1 and the stop frequency f2 is linearly reduced. Performs a process of making the rise and fall of the frequency characteristic “smooth”, such as linearly increasing the frequency between the stop frequency f3 and the cutoff frequency f4. It is possible to prevent inconvenience such that natural filtering is emphasized, and to obtain a natural sound output corresponding to the variably controlled frequency characteristic.

In general, the number of samples of audio data in the frequency domain is equal to the number of samples of audio data in the time domain. However, the multiplication processing can perform the operation processing more easily than the product-sum operation processing. . For this reason, when the filter unit 4 is realized by a high-speed arithmetic processing circuit such as a digital signal processor (DSP), the sound reproducing apparatus converts the sound data of the corresponding frequency resolution into a coefficient (filtering) corresponding to the set frequency characteristic. Coefficient) to control the frequency characteristic variably.
Can be reduced to several samples of audio data in the frequency domain, and the load on the DSP can be greatly reduced.

Further, when the filtering coefficient is formed, the normalization processing is performed with the frequency characteristic data of 0 dB as "1", so that the amount of calculation performed by the DSP can be reduced.

In the description of the first embodiment,
Although the frequency resolution of the audio data in the frequency domain is assumed to be constant in the entire frequency range, this may be set to a frequency resolution corresponding to each frequency range.

A plurality of frequency characteristic data are stored in the frequency characteristic setting section 5 in advance, and the user selects a desired frequency characteristic from the plural frequency characteristic data, which is externally input by the user. The frequency characteristic setting unit 5 may form frequency characteristic data in accordance with the frequency characteristics thus obtained, and supply this to the filter unit 4.

Further, the filter unit 4 normalizes the frequency characteristic data for each frequency resolution based on the frequency characteristic data of 0 dB. However, this is based on the frequency characteristic data of an arbitrary level. A conversion process may be performed.

The filter unit 4 performs both the formation of the filtering coefficient and the multiplication processing of the filtering coefficient and the audio data in the frequency domain. The filter unit 4 may perform only the multiplication process.

In the description of the first embodiment,
Although the MDCT coefficient is used as audio data in the frequency domain, audio data in another frequency domain such as a DCT coefficient may be used. When audio data in another frequency domain is used, when the audio data is converted to audio data in the time domain, the audio data in the time domain is converted using a conversion method corresponding to audio data in another frequency domain, such as inverse DCT processing. Needless to say, it is sufficient to convert the data into audio data.

Next, an audio reproducing apparatus according to a second embodiment of the present invention will be described.

In the audio reproducing apparatus according to the second embodiment, the filter unit 4 sets the mantissa part of the audio data in the frequency domain to “M”, the exponent part to “E”, and sets “M × 2 ＾ E ( M ×
expE) ”.

The audio reproducing apparatus according to the second embodiment operates in the same manner as the audio reproducing apparatus according to the first embodiment except that the filter unit 4 performs such an operation. Therefore, in the description of the audio reproducing apparatus according to the second embodiment, only the filter unit 4 will be described, and redundant description of other parts will be omitted.

That is, the inverse quantization unit 3 calculates “M × 2 ＾
The audio data in the frequency domain reproduced in the form of "E"
It consists of samples resolved at a fixed frequency resolution.

The filter unit 4 provided in the audio reproducing apparatus according to the second embodiment includes a frequency characteristic setting unit 5
Is sampled at the frequency resolution of the audio data in the frequency domain to form discrete frequency characteristic data (see FIG. 2B), and 0 dB of the discrete frequency characteristic data is set to a maximum value. , And-dB is set as a filtering coefficient of “−∞”. This filtering coefficient is added to the exponent “E” of the audio data in the frequency domain of the corresponding frequency, and a new processing is performed. Form an exponent "E '". Then, using this exponent part “E ′”, the audio data in the frequency domain of “M × 2 ＾ E” is filtered (see FIG. 2C).

As a result, when the filtered audio data in the frequency domain is converted into audio data in the time domain through the inverse converter 6 and supplied to the speaker device, the audio data is set by the frequency characteristic setting unit 5. In addition to obtaining an acoustic output having a desired frequency characteristic through the speaker device, the same effects as those of the audio reproducing device according to the above-described first embodiment can be obtained.

Finally, the description of the above-described embodiment is merely an example, so that the audio reproducing apparatus according to the present invention is not limited to this, and does not depart from the technical idea according to the present invention. Of course, various changes are possible within the range according to the design and the like.

[0066]

The audio reproducing apparatus according to the present invention reproduces the compressed and encoded audio data by changing it to a plurality of desired frequency characteristics without increasing the amount of calculation and the memory required for storing the filter coefficients. be able to.

[Brief description of the drawings]

FIG. 1 is a block diagram of an audio reproducing device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing a state where the frequency characteristic of audio data in a frequency domain is variably controlled in accordance with the frequency characteristic set in the audio reproducing apparatus according to the first embodiment.

FIG. 3 is a diagram illustrating a frequency characteristic of a low-pass filter set in the audio reproduction device according to the first embodiment.

FIG. 4 is a diagram illustrating frequency characteristics of a high-pass filter set in the audio reproduction device according to the first embodiment.

FIG. 5 is a diagram illustrating frequency characteristics of a band-pass filter set in the audio reproducing device according to the first embodiment.

FIG. 6 is a diagram showing frequency characteristics of a band rejection filter set in the audio reproduction device according to the first embodiment.

FIG. 7 is a block diagram of a conventional audio reproducing device.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Input terminal, 2 ... Unpacking part, 3 ... Inverse quantization part, 4
... Filter unit 5 ... Frequency characteristic setting unit, 6 ... Inverse conversion unit, 7 ... Output terminal

Claims (5)

[Claims] 1. A frequency characteristic setting unit for setting a desired frequency characteristic, and an externally supplied encoded audio information in a frequency domain supplied according to the frequency characteristic set by the frequency characteristic setting unit. Frequency characteristic variable means for variably controlling the frequency characteristic, and time domain audio forming means for converting the frequency-domain audio information variably controlled in frequency characteristic into time-domain audio information and outputting the time-domain audio information by the frequency characteristic variable means An audio playback device, comprising: 2. The frequency characteristic variable unit includes: a coefficient forming unit that forms a coefficient for variably controlling a frequency characteristic based on the frequency characteristic set by the frequency characteristic setting unit; 2. The audio reproducing apparatus according to claim 1, further comprising: a multiplying unit that varies a frequency characteristic of the audio information in the frequency domain by multiplying the obtained audio information in the frequency domain by the coefficient. 3. The coefficient forming unit forms a discrete frequency characteristic by sampling the frequency characteristic set by the frequency characteristic setting unit, and performs a normalization process based on a predetermined value of each sampled value. 3. The sound reproducing apparatus according to claim 2, wherein the coefficient for controlling the frequency characteristic is variably formed by performing the following. 4. The frequency characteristic setting means includes at least a low-pass filter and / or a high-pass filter.
4. The audio reproducing apparatus according to claim 1, wherein a band-pass filter and / or a band rejection filter can be set. 5. The frequency characteristic setting means sets a stop frequency higher or lower than a cutoff frequency when setting the frequency characteristic, and linearly sets a stop frequency from the cutoff frequency to the stop frequency. 5. The audio reproducing apparatus according to claim 4, wherein the frequency characteristic is set so as to change in a curved or stepwise manner.