close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JPH0646489

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JPH0646489
[0001]
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an
apparatus for reducing noise in an audio signal for reducing quantization errors (noises)
generated when quantizing an audio signal.
[0002]
2. Description of the Related Art Currently, as digital audio equipment that handles audio signals
in digital form, for example, a so-called compact disc (CD) player, a so-called digital audio tape
recorder (DAT), etc. exist. In these digital audio devices, various unified standards are defined. For
example, the bit length of digital audio signals handled by these devices is defined to 16 bits by
the unified standard. In addition, as digital audio signals in these digital audio devices, digital
audio signals obtained by encoding analog audio signals (sound waveform signals) by simple
linear quantization such as so-called PCM which is linear pulse coding may be used. It is used.
[0003]
By the way, in recent years, in the digital audio apparatus as described above, it is desired to
obtain reproduced sound of higher quality in terms of audibility than reproduced sound actually
obtained from the above-mentioned unified standard. In order to obtain such reproduced sound
better in terms of hearing, for example, it is considered effective to reduce noise components
10-04-2019
1
contained in digital audio signals handled by these digital audio devices. As described above, the
reproduced sound obtained from the audio signal in which the noise component is reduced is
better in terms of hearing with less noise.
[0004]
As a noise component reduction process of the digital audio signal, for example, a quantization
error reduction process by so-called error feedback performed at the time of quantization of the
audio signal is known. This is to feed back a quantization error (quantization noise, quantization
distortion) generated when the audio signal is quantized by the quantizer to the input side of the
quantizer through a noise filter.
[0005]
The noise reduction apparatus using such error feedback is described by the present applicant in
Japanese Patent Application Nos. 2-185552, 2-185555, and 2-185556. is suggesting.
[0006]
Here, quantization noise in linear quantization such as the above-mentioned PCM coding has flat
frequency characteristics over the entire frequency band of the audio signal.
However, human ears have different auditory sensitivities depending on the frequency of sound,
and quantization error reduction processing by error feedback is not always effective in terms of
auditory sense. Also, depending on the level of the input signal, quantization noise may be a
hearing problem.
[0007]
The present invention has been made in view of such circumstances, and it is an object of the
present invention to provide a noise reduction device which can effectively reduce the
quantization error (quantization noise) from the viewpoint of hearing. It is a thing.
[0008]
According to a noise reduction apparatus according to the present invention, an audio signal in
10-04-2019
2
which a quantization error generated by a quantizer is fed back to the input side of the quantizer
through a noise filter. In the noise reduction device of the present invention, level detection
means for detecting the level of the input signal, dither generation means for generating a dither
signal according to the output from the level detection means, and the output from the dither
generation means The problem described above is solved by including addition means for
adding.
[0009]
Here, the level detection means may supply an output signal to the dither generation means
when the input signal is at a level at which a sense of hearing is felt.
Further, the dither generation means may change the level of the dither in accordance with the
level of the input signal.
[0010]
The noise reduction apparatus according to the present invention further comprises a
quantization means for quantizing an input audio signal, and a quantization error (quantization
error by subtracting the input signal to the quantization means from the output signal of the
quantization means). Noise) extracting means, filter means (noise filter) to which the output from
the subtracting means is supplied, combining means such as an adder for combining the output
from the filter means and the input signal, and random Or dither generation means for
generating a pseudo random dither signal, dither control means for controlling the frequency
characteristic and level of the dither generation means, and means for controlling the
characteristics of the filter means, filter coefficient calculation means and equal loudness Filter
control means including data generation means for generating data relating to characteristics,
said data being supplied to said filter coefficient calculation means By comprising a, for solving
the above problems.
[0011]
Here, preferably, the dither control means further includes means for controlling the frequency
characteristic and the level of the dither signal based on the signal level of the input audio signal.
10-04-2019
3
Preferably, the filter control means further includes means for generating control data for
correcting the low frequency band component of the filter characteristic.
Furthermore, the filter control means preferably includes means for controlling the filter
characteristic according to the masking effect of the input audio signal.
[0012]
Furthermore, the filter control means implements a filter characteristic that implements noise
shaping such that an allowable noise spectrum obtained in consideration of the masking effect
on the frequency axis or on the time axis is provided according to the spectrum of the input
audio signal. It is preferable to generate a control signal to obtain. Furthermore, when the signal
level of the input signal is small, the filter control means takes into account the equal loudness
curve, and the allowable noise spectrum taking into account the masking effect as the signal level
increases becomes the value of the input audio signal. It is preferable to generate a control signal
to obtain filter characteristics for realizing noise shaping as given according to the signal level.
[0013]
In this noise reduction apparatus, the input signal is first adjusted by adding the dither signal,
and then the quantization error generated by the quantization means is input to the quantization
means through the filter means (so-called noise filter). Return to the side. The filter coefficients of
the noise filter are set based on the information related to the equal loudness curve according to
the human auditory characteristics. According to the present invention, noting that the equal
loudness curve corresponds to the auditory characteristics of the human ear, the quantization
error is a noise filter whose coefficient is set based on the information related to the equal
loudness curve. Since the feedback to the input of the quantizer is performed, the quantization
error in the frequency band easy to hear can be reduced. Therefore, it is possible to reduce
auditory noise and improve the auditory dynamic range.
[0014]
DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment to which a noise
10-04-2019
4
reduction apparatus according to the present invention is applied will be described with
reference to the drawings. In FIG. 1, a schematic configuration of a quantization error (noise)
reduction apparatus according to a first embodiment is shown in the form of a block diagram.
[0015]
In the embodiment of the quantization error (noise) reduction device shown in FIG. 1, the
quantization error generated by the quantizer 11 is fed back to the input side of the quantizer 11
through the noise filter 13. The filter coefficients of the noise filter 13 are set based on the
information related to the so-called equal loudness curve RC shown in FIG. 2 according to the
human auditory sense characteristic.
[0016]
Here, a digital audio signal obtained by sampling at an arbitrary sampling frequency is sent to
the input terminal 1 of FIG. The digital audio signal is added to the dither signal generated by the
dither generation circuit 21, requantized by the quantizer 11, and output from the output
terminal 2.
[0017]
The quantization error reduction device of this embodiment (and of the other embodiments
described below) consists of two subsystems: a dither system and a noise shaping system. These
subsystems act in a complementary manner and their properties can be varied independently.
[0018]
The dither subsystem comprises a level detector 20, a dither generation circuit 21 and an adder
10. The dither signal generated by the dither generation circuit 21 is added to the input signal by
the adder 10. The purpose of the dither subsystem is to generate a so-called dither signal,
defined as a random or pseudo-random signal, which is quantized such that the signal correlation
between the quantization error and the input signal is reduced It is added to the previous input
signal. The correlated error sounds better than the uncorrelated error, and reducing the
10-04-2019
5
correlation between this signal and the error reduces the auditory quantization noise. Also, this
decorrelation can enhance the effect of the noise shaping subsystem (described later) on the
input audio signal where the correlation between the signal and the error is significant.
[0019]
The dither generation circuit 21 generates a dither. The frequency characteristics of this dither
signal depend on the specific settings of this embodiment and are determined by the
characteristics of the class of input signal to be processed. The spectrum of a typical dither
consists of so-called white dither (equal at all frequencies), which forms a theoretically optimal
uncorrelated, and energy concentrated in the inaudible region of the signal spectrum. It contains
a more reduced so-called high frequency dither and an all zero dither which is suitable when the
input signal is already sufficiently random and does not need to be further decorrelated.
[0020]
The power spectrum of a non-dithered 1-bit amplitude sine wave is shown in FIG. 4 and the error
correlation is shown as a resonant spectrum peak. The same signal is shown in FIG. 5 as
quantized with white dither, indicating that there is a single spectral peak and that the
quantization error is not correlated to the signal. FIG. 6 shows the same signal quantized with
high frequency dither, with the dither power being high but concentrated in areas where it is not
noticeable to the ear, so the dither is less audible.
[0021]
Depending on the specific setting of this embodiment, the level detector 20 is used to control the
type of dither generated by the dither generation circuit 21 or the level of the dither or both. For
example, it is inappropriate to add dither to the input signal consisting of all zeros, so this
condition is detected by the level detector 20 and the dither generation circuit 21 is properly
controlled. In addition, low level input signals often require more dither than high level signals,
so the level of the dither gradually decreases as the signal level increases.
[0022]
10-04-2019
6
By the way, in the quantization error reduction device of this embodiment, the quantization error
is reduced by the noise shaping subsystem. The noise shaping subsystem subtracts the input to
the quantizer 11 from the output of the quantizer 11 to obtain an adder 12 that obtains a
quantization error generated in the quantization by the quantizer 11; A filter characteristic is set
by a filter coefficient to be described later, a noise filter 13 which performs filtering processing
on an output from the adder 12 and an adder which adds an output from the noise filter 13 to an
input of the quantizer 11 And so-called error feedback circuit. The error feedback circuit
performs quantization error reduction processing (so-called noise shaping processing).
[0023]
Furthermore, the quantization error reduction device of the present embodiment includes an
equal loudness curve generation circuit 15 for generating data of an equal loudness curve RC as
shown in FIG. 2 corresponding to human auditory characteristics, and the equal loudness curve
generation circuit 15 And a filter coefficient calculation circuit 14 that calculates the filter
coefficient of the noise filter 13 based on the output from the circuit of FIG.
[0024]
Here, the equal loudness curve RC in FIG. 2 corresponds to the human auditory characteristics.
This curve RC is obtained, for example, by connecting a curve segment of sound pressure for
each frequency that sounds as loud as the sound pressure of a pure tone of 1 kHz, which is the
loudness of the sound It is also called an equal sensitivity curve. According to the equal loudness
curve RC as shown in FIG. 2, the human hearing is sharp at around 4 kHz, and the sound
pressure is 8 to 10 dB lower than the 1 kHz sound pressure, and has substantially the same
magnitude (loudness) as 1 kHz. I can hear the sound. On the other hand, at 10 kHz, for example,
it is difficult to hear sound by about 20 dB as compared to around 4 kHz.
[0025]
Information (information of acceptable noise spectrum) related to the equal loudness curve RC
(or its approximate curve) is output from the equal loudness curve generation circuit 15 and sent
to the filter coefficient calculation circuit 14. Therefore, in the equal loudness curve generation
circuit 15, filter coefficients are calculated based on the information related to the equal loudness
10-04-2019
7
curve RC. The calculated filter coefficients are further sent to the noise filter 13. In this way, the
audibility dynamic range can be improved by noise shaping the audio signal by the error
feedback circuit using the noise filter 13 having the filter coefficient based on the information
related to the equal loudness curve RC. . That is, by performing noise shaping using an
acceptable noise spectrum (acceptable noise level) obtained in consideration of equal loudness
curve RC, noise shaping more effective for hearing can be derived, and this makes it possible to It
is possible to improve the dynamic range of the sense of hearing.
[0026]
Furthermore, in the present embodiment, when determining the filter characteristics of the noise
filter 13, so-called masking effects are considered. Here, the masking effect is a phenomenon in
which a certain signal is masked by another signal and can not be heard due to human auditory
sense characteristics. The masking effects include a masking effect (temporal masking) on signals
on the time axis and a masking effect (simultaneous masking) on signals on the frequency axis. If
noise is present in the part to be masked, the noise can not be heard by the masking effect. As a
result, when the method by the quantization error reduction processing in consideration of the
masking effect is executed, the audibility dynamic range is improved. In order to determine the
filter characteristics in consideration of the masking effect, for example, filter coefficients in
consideration of the masking effect on the frequency axis are set in advance in the filter
coefficient calculation circuit 14 of the present embodiment. For example, in order to cope with
normal audio speech including many intermediate and low frequency components, fixed filter
coefficients are set in consideration of the masking effect at low frequencies. Alternatively, to
obtain the ability to cope with the masking effects corresponding to the spectrum of the input
audio signal, techniques are used that generate adaptive filter coefficients according to the
spectrum.
[0027]
In this way, the filter coefficients from the filter coefficient calculation circuit 14 can be obtained
in consideration of the equal loudness curve RC and the masking effect. Therefore, the filter
characteristic of the noise filter 13 is set based on the fixed or adaptive filter coefficient in which
the masking effect is considered and the filter coefficient associated with the equal loudness
curve.
[0028]
10-04-2019
8
That is, the noise filter 13 at this time acts as a filter having a filter characteristic indicated by the
curve MR as shown in FIG. 3 obtained from the masking effect and the equal loudness curve. By
providing the noise filter 13 with the filter characteristic indicated by the curve MR in FIG. 3, the
quantization error spectrum sent to the noise filter 13 changes in accordance with the curve MR.
By adding the output from the noise filter 13 to the input audio signal, the quantization error in
the quantizer 11 is reduced (based on noise shaping). Here, in the curve MR of FIG. 3, when the
equal loudness curve RC of FIG. 2 is taken into consideration, it is conceivable to increase the
response in the frequency band lower than 4 kHz (that is, the tolerance noise may also be
increased). The filter characteristics in the frequency band lower than 4 kHz are flat (flat) as in
the second embodiment described later. The reason why such a mechanism works will be
described next.
[0029]
The 4 kHz or less of the equal loudness curve RC has a steep change despite the fact that the
bandwidth is not wide, so creating a noise filter 13 that matches the equal loudness curve RC of
4 kHz or less increases the dimension of the filter It is because it When the dimension of the
filter is increased in this manner, the configuration becomes complicated and enlarged. However,
at this time, since an effect commensurate with the scale of the filter can not be obtained, in the
second embodiment described below, the filter characteristic of 4 kHz or less is flat as described
above.
[0030]
That is, FIG. 7 shows a second embodiment of the present invention, and the blocks
corresponding to the respective parts in FIG. In FIG. 7, a low frequency band correction control
signal generation circuit 16 is further provided, and the low frequency band correction signal
generated by this circuit is sent to the filter coefficient calculation circuit 14. Therefore, flat (flat)
filter characteristics are realized in the low frequency band as shown by the curve MR in FIG.
Furthermore, the low frequency band correction signal is such as to be formed taking into
account the masking effect. In general, the middle and high frequency bands are often used in
normal audio sound, and the above-described masking effect in the middle and high frequency
bands of the audio sound is effective. Therefore, in the noise filter 13 of this embodiment, the
response (response) of the curve MR of the filter coefficient of FIG. 3 is not reduced to such an
extent that the response of the equal loudness curve RC of FIG. Made more gentle compared to
10-04-2019
9
the curve). That is, in order to realize this, as described above, the filter coefficient in which the
masking effect is considered is set. The filter characteristic of the noise filter 13 is set as shown
in FIG. 3, and the frequency characteristic of the quantization noise after the quantization noise
processing is performed using the actual audio speech is shown in FIG.
[0031]
A third embodiment using the masking effect will be described with reference to FIG. Also in FIG.
9, the blocks corresponding to those in FIG. 1 are given the same reference numerals.
[0032]
The quantization noise reduction device of FIG. 9 is configured to feed back the quantization
error generated by the quantizer 11 to the input side of the quantizer 11 via the noise filter 11.
More specifically, this quantization noise reduction device includes a level detector 19 that
detects the level of an input audio signal, a frequency analysis circuit 17 that analyzes the
frequency of the input audio signal for each critical band, and a human being. And the output
from the frequency analysis circuit 17 and the output from the equal loudness curve generation
circuit 15 as shown in FIG. And a tolerable noise spectrum calculation circuit 18 which calculates
the tolerable noise spectrum based on the obtained synthetic information. In this quantization
noise reduction device, the filter characteristics of the noise filter 13 are set based on the output
information of the allowable noise spectrum calculation circuit 18.
[0033]
That is, in the quantization error reduction device of the present embodiment, the quantization
error generated at the time of quantization in the quantizer 11 is subtracted by subtracting the
input to the quantizer 11 from the output of the quantizer 11. A so-called error is generated by
the adder 12 to be obtained, the noise filter 13 for filtering and outputting the output of the
adder 12, and the adder 10 for adding the output of the noise filter 13 to the input side of the
quantizer 11. A feedback circuit is configured. Here, the filter characteristics of the noise filter 13
are determined as follows. In practice, the filter coefficient calculation circuit 14 calculates a
filter coefficient based on the information of the allowable noise spectrum described later of the
allowable noise spectrum calculation circuit 18 and sends the filter coefficient information to the
noise filter 13.
10-04-2019
10
[0034]
Therefore, in the above-described error feedback circuit, a quantization error reduction process
(so-called noise shaping process) based on an allowable noise spectrum described later is
performed. The signal thus processed is output from the output terminal 2.
[0035]
By the way, when performing quantization error reduction processing (noise shaping processing)
of an audio signal by the above-described error feedback circuit, the processing in consideration
of so-called masking of the input signal spectrum is performed, thereby expanding the dynamic
range on audibility. be able to. By performing processing in consideration of this masking effect,
for example, noise shaping in accordance with the spectrum of the input audio signal in which
the pattern of the signal spectrum is fixed to some extent, ie, so-called masking effect of the input
audio signal spectrum will be obtained. Noise shaping using the specified allowable noise
spectrum can be mentioned.
[0036]
Alternatively, when there is a change in the spectrum of the input audio signal, there is noise
shaping using an acceptable noise spectrum adaptive to the change in spectrum obtained in
consideration of the masking of the spectrum. Here, the masking effect is a phenomenon in
which a certain signal is masked by another signal and can not be heard due to human auditory
sense characteristics. The masking effects include a masking effect (temporal masking) on signals
on the time axis and a masking effect (simultaneous masking) on signals on the frequency axis.
Due to this masking effect, even if there is noise in the portion to be masked, this noise can not
be heard. For example, as for the masking effect at the same time, as shown in FIG. 10, when the
frequency response of the signal S of a certain frequency is 0 dB, the masking effect acts on the
curve M (about −15 dB or less) by the signal S. Become.
[0037]
10-04-2019
11
Furthermore, it is divided into so-called critical bandwidths (critical bands) using human auditory
characteristics, and divided at this critical bandwidth, and frequency analysis is performed for
each of the critical bands. As the division at the critical bandwidth at this time, for example, the
input audio signal is converted to a component on the frequency axis by, for example, fast
Fourier transform (FFT), and then the amplitude term Am (m = 0 to 0) of this FFT coefficient.
1024) is divided into, for example, a group Gn of 25 bands (n indicates the number of each band,
n = 0 to 24) in the above critical bandwidth where the bandwidth becomes wider as the high
frequency region considering human auditory characteristics It can be divided into bands.
[0038]
Further, the frequency analysis for each critical band can be obtained, for example, by taking the
sum (peak of peak amplitude Am, average or total energy) of each amplitude term Am for each
band according to the following equation (1) An analysis can be performed to determine the socalled Burk spectrum (spectrum of the sum) Bn. Bn = 10 log 10 Cn (Pn) 2 (1)
[0039]
Here, n is 0 to 24, Cn is the number of elements in the nth band (band), that is, the amplitude
term (number of points), and Pn is a peak value of each band. The bark spectrum Bn of each
band is shown, for example, in FIG. In the example of FIG. 11, in order to simplify the drawing,
the total number of bands is expressed by 12 bands (B1 to B12). The frequency analysis circuit
17 performs the division at the critical bandwidth and the frequency analysis for each band, and
the output information is sent to the allowable noise calculation circuit 18.
[0040]
Information of the equal loudness curve RC is generated and output from the equal loudness
curve generation circuit 15. That is, by performing noise shaping using an allowable noise
spectrum obtained in consideration of equal loudness curves, more effective noise shaping can
be performed on hearing. Thus, the audibility dynamic range of the reproduced sound can be
improved. Information on the equal loudness curve RC (or its approximate curve) is output from
the equal loudness curve generation circuit 15 and sent to the allowable noise spectrum
calculation circuit 18.
10-04-2019
12
[0041]
Therefore, in the allowable noise spectrum calculation circuit 18, the allowable noise spectrum is
calculated based on the output information from the equal loudness curve generation circuit 15
and the output information from the frequency analysis circuit 17. At this time, convolution is
performed from the Bark spectrum Bn in the frequency analysis circuit 17 in consideration of the
influence between the bands using the following equation (2) (convoluting a predetermined
weighting function), Convoluted Bark spectra Sn for each band are calculated.
[0042]
Sn = Hn * Bn (2) where Hn is a coefficient of convolution. By this convolution, the portion
indicated by the dotted line in FIG. 11 is summed. Furthermore, using this convolution processed
Bark spectrum Sn and the required S / N value On (N = 0 to 24), convolution processing is
performed according to the following equations (3) and (4) The calculated masking threshold Tn
is calculated.
[0043]
On = N−K × n (3) Tn = Sn−On (4) For example, when N is 38, K can be 1. At this time, the
result is that there is little deterioration in sound quality. That is, as shown in FIG. 12, the sound
having an intensity equal to or lower than the level indicated by each level of the convolution
processed masking threshold Tn is masked. Thereafter, the deconvoluted masking threshold Tn
is deconvoluted using the following equation (5) to calculate a tolerable noise level (permissible
noise spectrum) TFn. In practice, for example, a direct current (DC) gain Dn of convolution based
on the coefficient Hn is subtracted.
[0044]
TFn=Tn−Dn ・・・(5)
[0045]
The allowable noise spectrum calculation circuit 18 is obtained by combining the output
information from the frequency analysis circuit obtained as described above and the output
10-04-2019
13
information from the equal loudness curve generation circuit 15 described above. The allowable
noise spectrum is determined based on the combined information.
[0046]
Here, the allowable noise level in the allowable noise spectrum based on the above-mentioned
equal loudness curve RC may be lower than the allowable noise level on which the masking effect
acts depending on the level of the input audio signal.
That is, for example, when the level of the input audio signal is high, the level of the allowable
noise spectrum based on the equal loudness curve may be simultaneously masked by the
allowable noise level to which the masking effect by the input audio signal acts. is there.
[0047]
From the above viewpoint, in the present embodiment, the level of the input audio signal is
detected by the level detector 19, and based on the level detection output, the output information
from the equal loudness curve generation circuit 15 and the frequency analysis circuit 17 It
changes the composition ratio with the output information from.
Here, the synthesis of the output information from the equal loudness curve generation circuit
15 and the frequency analysis circuit 17 is performed, for example, for each frequency band. In
this case, level detection by the level detector 19 is performed for each band. Therefore, the
combination ratio can be changed for each band based on the level detection output for each
band. That is, with regard to synthetic information for obtaining the noise spectrum in the
allowable level spectrum calculation circuit 18, for example, when the low level of the input
audio signal is high and the masking effect in the low area is large, the low band is high. The
synthesis information is prepared at a synthesis ratio such that an acceptable noise spectrum
having high levels at low levels is obtained. Conversely, for example, when the high frequency
level is high and the masking effect in the high frequency region is large, the combined
information is obtained at a synthesis ratio such that an acceptable noise level such that the high
frequency region is high and the low frequency region is low. Made. The information of the
allowable noise spectrum thus obtained is sent to the filter coefficient calculation circuit 14.
Then, a filter coefficient corresponding to the allowable noise spectrum is output from the filter
coefficient calculation circuit 14 and sent to the noise filter 13.
10-04-2019
14
[0048]
From the above, it is assumed that the filter characteristic of the noise filter 13 corresponds to
the filter coefficient based on the allowable noise spectrum obtained by varying the synthesis
ratio for each band according to the level of the input audio signal. Become. Here, for example,
when the level of the input audio signal is flat, it is assumed that the filter characteristic of the
noise filter 13 is as shown by a curve MR in FIGS. At this time, when the input audio signal
becomes, for example, a signal S1 of a slightly high level in the low band as shown in FIG. 13, the
above-mentioned filter characteristics are changed by changing the synthesis ratio as described
above. The characteristic is as shown by the curve MR1 in FIG. 13 in which the low band of the
curve MR is slightly raised and the high band is slightly lowered.
[0049]
Further, for example, when the input audio signal is a low level signal S2 as shown in FIG. 14, the
filter characteristic of the noise filter 13 is such that the low range of the curve MR is largely
increased and the high frequency range is high. Is changed to a characteristic as shown by a
curve MR2 in FIG. Conversely, in the case where the input audio signal is a signal S3 of a slightly
higher level in the high region as shown in FIG. 15, the filter characteristic is such that the low
region of the curve MR is slightly lowered and the high region is slightly 15 is changed to a
characteristic as shown by the curve MR3 in FIG. 15 and, for example, when it is a signal S4
having a large level in the high region as shown in FIG. Is greatly lowered and the high region is
greatly increased, as shown by the curve MR4 in FIG. As shown in FIG. 13 to FIG. 16 described
above, by changing the filter characteristics, it is possible to perform noise shaping more
according to human auditory characteristics.
[0050]
That is, in other words, in the apparatus of this embodiment, when the level of the input audio
signal is small, the filter characteristic of the noise filter 13 is made to be a characteristic like the
equal loudness curve RC to perform noise shaping. . In addition, as the signal level increases, the
characteristics of the noise filter 13 are made flat according to the signal level of the input audio
signal in order to make the quantization noise less noticeable by the level of the input audio
signal. Furthermore, when the signal level is small, the noise filter 13 brings characteristics like
10-04-2019
15
the equal loudness curve RC closer to flat according to the signal level, and changes it to noise
shaping characteristics (masking characteristics etc.) according to the signal characteristics.
Make it That is, the characteristic of the noise filter 13 is a filter characteristic such as the equal
loudness curve RC when the signal level is small, and is a filter characteristic in which the
masking effect is considered when the signal level is large.
[0051]
In the curve MR showing the filter characteristics when the level of the input audio signal in FIG.
13 to FIG. 16 is flat, it is conceivable to increase the level of 4 kHz or less in consideration of the
equal loudness curve RC in FIG. (That is, it can be made to increase acceptable noise), but since
the change is steep despite the fact that the 4 kHz or less of this equal loudness curve RC is not
wide bandwidth, the equal loudness curve RC of this 4 kHz or less If the combined noise filter 13
is created, the dimension of the filter is increased. Thus, increasing the dimension of the filter
makes the configuration more complicated and larger. However, at this time, since an effect
commensurate with the scale of the filter can not be obtained, in the present embodiment, as
described above, the filter characteristic of 4 kHz or less is flat. Further, in the noise filter 13 of
the present embodiment, the masking effect in the low-pass of the audio voice is normally
applied as described above in the middle and low-pass areas of high frequency of use in the
audio voice. The 16 curves MR are not lowered as much as the equal loudness curve RC in FIG. 4
(the curve MR is a gentler curve than the equal loudness curve RC). That is, in order to do this, as
described above, the filter coefficient in consideration of the masking effect is set.
[0052]
FIG. 17 shows a specific example of a system configuration in which the device of this
embodiment is used in, for example, a system of encoders and decoders in a so-called compact
disc (CD). In FIG. 17, an analog audio signal is supplied to the input terminal 31. The analog
audio signal is converted into a 20-bit digital signal by an A / D converter 32, and then sent to a
20-bit compatible encoder 33 in which the quantization error reduction device of this
embodiment is incorporated. The quantization error reduction processing is performed by the
encoder 33 and the data encoded into 16-bit data is recorded on the CD. The data recorded on
the CD is converted into an audio signal by the reproduction circuit 34 and the D / A converter
35 of the existing CD player, and output from the output terminal 36 for reproduction. That is,
since the data recorded on the CD has the quantization error reduced by the quantization error
reduction device of this embodiment, the sound obtained by reproducing the CD has a dynamic
range over the sense of hearing. It will be expensive.
10-04-2019
16
[0053]
Further, FIG. 18 shows a specific example of a system configuration via a medium for recording
data with, for example, 10 bits unlike the above-mentioned CD. In this case, the analog signal
supplied to the input terminal is, for example, a 10-bit compatible encoder 43 in which the signal
in which the digital data of 16 bits is converted by the A / D converter 42 is built in Sent to The
quantization error reduction processing is performed by the encoder 43, and the data encoded
into 10-bit data is recorded on the medium. The data recorded on the medium is converted into
an analog signal by the reproduction circuit 44 and the D / A converter 45 of the existing
reproduction machine and output from the output terminal 46. Also in this case, the audibility
dynamic range of the obtained reproduced signal is increased.
[0054]
FIG. 19 shows a specific example of using the apparatus of this embodiment in a system of D / A
conversion that performs oversampling. In this case, the analog signal supplied to the input
terminal 51 is converted to, for example, 20-bit digital data by the D / A converter 52 that
performs oversampling, and the quantization error reduction device 53 of this embodiment is
transmitted through the transmission path. Sent to A quantization error reduction process is
performed by the quantization error reduction device 53, and the D / A converter 54 converts
the signal into an analog signal and outputs the analog signal from the output terminal 55. As a
result, oversampling can be performed, the resolution of D / A conversion can be reduced, and
the D / A converter 54 can be easily formed in a direction to increase the linearity accordingly.
[0055]
The present invention is not limited to the above embodiments, and it is needless to say that
various additions and modifications can be made without departing from the scope defined and
guaranteed by the claims of the present invention. .
[0056]
As apparent from the above description, according to the noise reduction device of the present
invention, the quantization error generated by the quantizer is fed back to the input side of the
quantizer through the noise filter. In the noise reduction device for an audio signal, the dither
signal generated according to the level of the input signal is added to the above input signal,
10-04-2019
17
whereby the quantization noise becomes an auditory problem. The noise becomes hard to hear,
and the quantization noise does not become a problem in hearing.
[0057]
Also, by setting the filter coefficient of the noise filter in consideration of the equal loudness
curve and the masking effect, for example, the combination ratio of the information of equal
loudness curve and the frequency analysis information of the input audio signal corresponds to
the level of the input audio signal. By setting based on information obtained by being changed, it
is possible to reduce auditory noise and improve the auditory dynamic range.
[0058]
Furthermore, by applying the noise reduction device according to the present invention to, for
example, a digital audio device of the existing unified standard, the reproduced sound of the
dynamic range expanded in hearing sense more than the dynamic range actually obtained from
the unified standard. You will be able to get it.
That is, it is possible to improve the audibility dynamic range of the reproduction audio signal by
using the same reproduction apparatus as in the prior art while maintaining the unification
standard without adding any change to the unification standard.
10-04-2019
18
Документ
Категория
Без категории
Просмотров
0
Размер файла
32 Кб
Теги
description, jph0646489
1/--страниц
Пожаловаться на содержимое документа