JP3341440B2 - Information encoding method and apparatus, information decoding method and apparatus, and information recording medium - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information encoding method which encodes input digital data by so-called high-efficiency encoding, and transmits, records, reproduces and decodes the digital data to obtain a reproduced signal. The present invention relates to an apparatus, an information decoding method and apparatus, and an information recording medium.

[0002]

2. Description of the Related Art Hitherto, various methods for highly efficient encoding of a signal such as audio or voice have been known. For example, an audio signal or the like on a time axis is divided into a plurality of frequency bands without being blocked. Division coding (sub-band coding: SBC), which is a non-blocking frequency band division method for performing coding, and converting a signal on the time axis into a signal on the frequency axis (spectral conversion) to generate a plurality of frequency bands , And a coding frequency band division method for encoding for each band, so-called conversion encoding. Further, a high-efficiency coding method combining the above-described band division coding and transform coding is also considered.In this case, for example, after performing band division by the band division coding, The spectrum of the signal for each band is converted into a signal on the frequency axis, and coding is performed for each band that has been subjected to the spectrum conversion.

Here, as a band dividing filter used in the above-mentioned band dividing coding, for example, QM
There is an F filter, and this QMF filter is described in the document "Digital Coding of Speech in
Subbands "(Digital coding of speech in subbands,
RECrochiere, Bell Syst.Tech.J., Vol.55, No.819
76).

[0004] Also, a document "Polyphase Quadrature Filters-New Band Division Coding Technology" (PoPo
lyphase Quadrature filters -A new subband coding t
echnique, Joseph H. Rothweiler, ICASSP 83 BOSTON)
Describes an equal bandwidth filter division technique.

[0005] The above-mentioned spectral conversion includes:
For example, an input audio signal is divided into blocks in a predetermined unit time (frame), and a discrete Fourier transform (DFT), a cosine transform (DCT), a modified DCT transform (MDCT), or the like is performed for each block, so that the time axis is changed to the frequency axis. There is a spectrum conversion to be performed. The above MDCT is described in the document “Time domain aliasing.
Subband / Transform Coding Using Cancellation-Based Filter Bank Design "(Subband / TransformCoding
Using Filter Bank Designs Based on Time Domain Ali
asing Cancellation, JPPrincen, ABBradley, Uni
v. of Surrey, Royal Melbourne Inst. of Tech. ICASSP
1987).

As described above, by quantizing the signal divided for each band by the filter or the spectrum conversion, the band in which the quantization noise occurs can be controlled.
Aurally more efficient encoding can be performed by utilizing properties such as a masking effect. Further, if the normalization is performed for each band, for example, with the maximum value of the absolute value of the signal component in that band before the quantization is performed, more efficient encoding can be performed.

Here, as a frequency division width when quantizing each frequency component divided into frequency bands, for example, a bandwidth in consideration of human auditory characteristics is often used. That is, an audio signal may be divided into a plurality of bands (for example, 25 bands) in a band called a critical band (critical band) in which the band becomes generally wider as the frequency becomes higher. When encoding data for each band at this time, a predetermined bit distribution for each band or
Encoding by adaptive bit allocation (bit allocation) is performed for each band. For example, when the coefficient data obtained by the MDCT processing is encoded by the bit allocation, the MD of each block is encoded.
The MDCT coefficient data for each band obtained by the CT processing is encoded with an adaptive number of allocated bits. The following two methods are known as bit allocation methods.

For example, the document "Adaptive Transform Coding of Speech Signals.
R. Zelinski, P. Noll, IEEE Transactions of Accoustic
s, Speech, and Signal Processing, vol.ASSP-25, No.
4, August 1977), bits are assigned based on the magnitude of the signal for each band. In this method, the quantization noise spectrum is flattened and the noise energy is minimized, but the actual noise sensation is not optimal because the masking effect is not utilized from the auditory point of view.

Further, for example, in the document “Critical band encoder—
Digital Encoding of Perceptual Requirements of the Auditory System (The critical band coder --digital encodi
ng of the perceptual requirements of the auditory
system, MAKransner MIT, ICASSP 1980) describes a method of obtaining a required signal-to-noise ratio for each band and performing fixed bit allocation by using auditory masking. However, in this method, even when the characteristic is measured with a sine wave input, the characteristic value is not so good because the bit allocation is fixed.

[0010] In order to solve these problems, all bits that can be used for bit allocation include a fixed bit allocation pattern predetermined for each small block and a bit allocation depending on the signal size of each block. A high-efficiency coding apparatus has been proposed in which a division ratio is used to perform the division, the division ratio is made dependent on a signal related to an input signal, and the division ratio into the fixed bit allocation pattern is increased as the spectrum of the signal is smoother. ing.

According to this method, when energy is concentrated on a specific spectrum such as a sine wave input, a large number of bits are allocated to a block including the spectrum, thereby significantly improving the entire signal-to-noise characteristic. can do. In general, human hearing is extremely sensitive to signals having steep spectral components. Therefore, using such a method to improve the signal-to-noise characteristics merely improves the numerical values measured. But not
This is effective for improving sound quality in terms of hearing.

Numerous other methods have been proposed for the bit allocation method. Further, models relating to hearing are refined, and if the capability of the coding device is increased, more efficient coding can be perceived perceptually. Will be possible.

When the above-described DFT or DCT is used as a method of converting a waveform signal into a spectrum, M independent real number data can be obtained by performing conversion using a time block including M samples. In order to reduce the connection distortion between time blocks, normally, since the overlapped by sample blocks and _one M each neighboring, on average, in DFT or DCT (M-M ₁₎ with respect to samples M pieces of real number data are quantized and encoded.

On the other hand, when the above-mentioned MDCT is used as a method for converting into a spectrum, 2M samples overlapping N times each with the time on both sides are used.
Since M independent real number data are obtained, on average, MD
In CT, M real number data is quantized and encoded for M samples. In the decoding apparatus, it is possible to reconstruct a waveform signal by adding the waveform elements obtained by performing inverse transform in each block from the code obtained by using the MDCT while causing the blocks to interfere with each other. it can.

In general, by extending the time block for the conversion, the frequency resolution of the spectrum is increased and the energy is concentrated on a specific spectral component. Therefore, by using a MDCT in which the transform is performed with a long block length by overlapping the adjacent blocks by half each and the number of obtained spectral signals does not increase with respect to the number of original time samples, DFT or It is possible to perform more efficient coding than when DCT is used. In addition, by providing a sufficiently long overlap between adjacent blocks, distortion between blocks of a waveform signal can be reduced.

As described above, when a method of once decomposing a signal into frequency components and quantizing and encoding the frequency components is used, a waveform signal obtained by decoding and synthesizing the frequency components is also quantized. Noise is generated, but if the original signal component changes rapidly, the quantization noise on the waveform signal will be large even in areas where the original signal waveform is not large, and this quantization noise Since it is not concealed by masking, it becomes an obstacle to hearing. The quantization noise generated in this way in the attack part where the sound suddenly increases is called pre-echo.

In particular, when the signal is decomposed into a large number of frequency components using spectral conversion, the time resolution becomes poor,
Pre-echo occurs over a long period.

Here, the operation principle of the generation of the pre-echo when the spectrum conversion is used in the band division will be described.
This will be described with reference to FIG.

When quantization noise is added to the spectrum signal obtained by performing forward spectrum conversion on the input waveform signal SW using the window function or the window function shown in FIG. 7A, the quantization noise is added. Performing inverse spectrum conversion on the spectrum signal and returning it to the waveform signal on the time axis again,
The quantization noise spreads over the entire transform block.
Here, when the input signal waveform suddenly increases in the middle of the conversion block as in (B), the quantization noise QN becomes larger than the signal waveform SW in a section where the original signal waveform is small. As a result, simultaneous masking does not work, and the pre-echo causes hearing damage.

Here, if the conversion length of the spectrum conversion is shortened, the period during which the above-mentioned quantization noise is generated is also shortened. However, if so, the frequency resolution is deteriorated, and the coding efficiency in the quasi-stationary part is deteriorated. As a means for solving such a problem, there has been proposed a method of shortening the conversion length at the part where the signal waveform changes abruptly at the expense of frequency resolution.

FIG. 8 is a diagram for explaining an example of the prior art devised to reduce the hearing impairment due to such a pre-echo. Generally, for a quasi-stationary signal waveform, the longer the transform block length is, the more the energy concentrates on a specific spectral coefficient, so that the coding efficiency becomes higher, but the sound volume changes abruptly. If the conversion block length is long in the part, the above-mentioned pre-echo becomes a problem.

Therefore, the portion where the loudness of the sound changes rapidly, for example, the input signal waveform SW shown in FIG.
Where the amplitude of increases sharply, a short conversion window function or a short conversion window function that shortens the conversion block length is applied as shown in FIG.
As a result, if the period of occurrence of the pre-echo is sufficiently shortened, so-called reverse masking by the original signal is effective, so that there is no obstacle to hearing. In the method of FIG. 8, utilizing this, the conversion block length is selectively switched according to the properties of each part of the signal waveform.

Using this method, sufficient frequency resolution is ensured in the quasi-stationary part, and the pre-echo in the attack part has a sufficiently short generation period and is concealed by so-called reverse masking, so that efficient coding is performed. It becomes possible.

However, in the method of making the conversion length variable as described above, it is necessary to provide conversion means corresponding to conversions of different lengths in the encoding device and the encoding device. Furthermore, in this method, since the number of spectral components obtained by the conversion is proportional to the length of the conversion length, the frequency band corresponding to each spectral component also differs depending on the conversion length, and a plurality of spectra are grouped, for example, for each critical bandwidth. However, when trying to perform encoding, the number of spectra included in each critical band also differs, and the encoding and decoding processes become complicated. As described above, the method of making the conversion length variable has a disadvantage that both the encoding device and the decoding device are complicated.

As a method for solving the above-described pre-echo problem while keeping the conversion block length constant, Japanese Patent Application Laid-Open No. 3-132228 discloses an adaptive gain control for an input waveform signal. Later, DFT and D
A method is described in which a signal is converted into a spectrum signal using CT and coding is performed. Here, the gain control means increasing the gain (amplifying the amplitude) where the power level is small.

In this method, the coding apparatus performs gain control in which the gain is sharply reduced in the attack section before performing conversion to the spectrum signal, and increases the gain again in accordance with the attenuation in other sections than the attack section. The gain control is performed, and the decoding apparatus outputs a signal obtained by performing inverse gain control on the signal waveform obtained by performing the inverse spectrum conversion to correct the gain control. In this way, the quantization noise in the small-amplitude signal portion where the masking level becomes low is suppressed. Further, since the conversion length can always be kept constant, the configurations of the encoding device and the decoding device can be simplified.

However, in this method, it is necessary to perform gain control even when the signal is attenuated. Generally, gain control distorts the original signal waveform,
When converted to a spectrum, the energy distribution is dispersed, making it difficult to perform efficient coding. In particular, when the signal is attenuated, the forward masking that masks the sound that occurred before the previous sound works effectively, so it is more important to reduce the noise level itself than to control the generation of quantization noise temporally. It is. Also, it is not preferable to always perform the gain control process from the viewpoint of the amount of calculation processing.

As another method for preventing pre-echo while keeping the conversion block length constant, for example, Japanese Patent Laid-Open No.
Techniques such as those disclosed in JP-A-01526 and JP-A-63-7023 are known. In these publications, an encoding device cuts out an input signal waveform for each time block, applies a window, detects an attack portion, amplifies a small-amplitude waveform immediately before the attack portion, and then performs DFT or DC
T is converted into a spectrum signal using T, and the decoded signal is decoded.
(Inverse DFT: IDFT) and inverse DCT (Inverse DFT)
There has been proposed a method for preventing a pre-echo by performing an inverse transformation such as CT: IDCT) and then performing a process of correcting that a signal immediately before an attack unit is amplified by an encoding device. Also in this case, the conversion length can always be constant, and the configurations of the encoding device and the decoding device can be simplified.

Here, FIG.
No. 26 and Japanese Patent Application Laid-Open No. 63-7023 describe the principle of operation for encoding and decoding using the windowing processing technique. FIGS. 10 and 11 show this technique. 3 shows the flow of processing performed by the encoding device and the decoding device.

The input terminal 400 shown in FIG.
(A) of FIG. 9 is input, and the window circuit 401 sets a time window that is temporally continuous and superimposed on each other, and cuts out a time waveform signal as shown in FIG. A window function (characteristic curve referred to in the above-mentioned Japanese Patent Application Laid-Open No. 61-201526) is multiplied. The attack part detection circuit 402 detects a part (attack part) where the amplitude of the input signal sharply increases. The gain control circuit 403 performs processing so as to amplify the minute amplitude portion if an attack part is detected, and does not perform amplification processing if the attack part is not detected. An output from the gain control circuit 403 is sent to a forward spectrum conversion circuit 404, and DFT, D
It is converted into a spectrum signal by CT or the like. The spectrum signal thus obtained is normalized and quantized by the normalization / quantization circuit 405, then encoded by the encoding circuit 406, and extracted from the output terminal 407 as a code string.

In the decoding apparatus shown in FIG. 11, the code string signal supplied to the input terminal 410 is
In the decoding circuit 411, the inverse decoding of the encoding in the encoding circuit 406 is performed, and the inverse normalization / inverse quantization circuit 412 is applied.
Sent to An output from the inverse normalization / inverse quantization circuit 412 is output to an inverse spectrum conversion circuit 413 by IDFT or IDCT.
After the inverse conversion to the time domain is performed by the above-described method, the data is sent to the gain control correction circuit 414, and the gain control processing performed by the encoding device is corrected. The output from the gain control correction circuit 414 is sent to the adjacent block synthesis circuit 415, where the output from the gain control correction circuit 414 is synthesized with the adjacent block, and extracted through the output terminal 416.

In this method, after the window function is applied as described above, the process of detecting the attack portion is performed on the deformed waveform signal, so that the large-amplitude portion is mitigated at both ends of the block. For example, as shown in FIG. 9, an attack portion may not be detected in the block BL1 and an attack portion may be detected only in the next block BL2. However, the above-described DFT or DCT is used for spectrum conversion. In this case, if the spectrum obtained by performing the forward spectrum transformation is subjected to the inverse spectrum transformation, the original time-series block is completely restored.
If the decoding apparatus performs the gain control correction processing for each block, no problem occurs.

[0033]

However, in the method described in the embodiment of the method using the gain control, the gain control amount in the attack portion is fixed, and the predetermined amount depends on whether the attack portion is detected or not. It is determined whether or not to perform the gain control of the value of. In pre-echo, quantization noise generated in the frequency domain is converted to the time domain.Therefore, if the compression rate of coding increases and the quantization noise in the frequency domain increases, sound quality degradation due to pre-echo increases, but on the other hand, Performing more gain control than necessary spreads the energy distribution in the frequency domain, which is not desirable for efficient compression. Therefore, in the conventional method of alternately selecting whether or not to perform a predetermined value of gain control based on the presence or absence of detection of an attack portion, sound quality is prevented from being deteriorated particularly when the compression ratio is high. It was difficult.

The present invention has been made in view of such circumstances, and enables gain control in accordance with the degree of amplitude change of an attack portion, and enables more efficient encoding and decoding with higher sound quality. Alternatively, it is an object of the present invention to provide an information encoding method and apparatus, an information decoding method and apparatus, and an information recording medium that can perform recording and have a simple configuration and can effectively prevent pre-echo.

[0035]

According to the present invention, in order to solve the above-mentioned problems, the gain control amount of the attack section is changed in accordance with a change in the level of a waveform signal, so that the present invention is effective even when the compression ratio is high. This is to prevent pre-echo.

That is, in the information encoding method according to the present invention, there are provided a band division process for dividing an input waveform signal into a plurality of bands, a gain control process for controlling a gain of the band-divided waveform signal, and a gain control process. An information code for performing a frequency component decomposition process of blocking the processed input waveform signal for each predetermined unit time and decomposing it into frequency components, and an encoding process of output information of the frequency component decomposition process and control information of the gain control. The gain control process is performed for each band subjected to the band division process, and the control information of the gain control includes flag information indicating whether to perform the gain control, and performs the actual gain control. It is characterized in that the control information of a non-existing block comprises only the flag.

Further, the information coding apparatus according to the present invention comprises: a band division processing means for dividing an input waveform signal into a plurality of bands; a gain control processing means for performing a gain control processing of the band-divided waveform signal; Frequency component decomposition processing means for blocking an output signal from the gain control processing means for each predetermined unit time and decomposing the signal into frequency components; and encoding output information from the frequency component decomposition processing means and control information for gain control. And an encoding means for performing the processing, wherein the gain control processing is performed for each of the bands subjected to the band division processing, and the control information of the gain control indicates whether to perform gain control. It is characterized in that the control information of a block that includes flag information and does not perform actual gain control is composed of only the flag.

Next, an information decoding method according to the present invention provides a decoding process of a frequency component signal and gain control correction information, and a waveform signal of a plurality of bands which are blocked from the frequency component signal at predetermined unit times. Information decoding that performs a waveform signal generation process of generating a waveform signal, a gain control correction process of an output waveform signal of the waveform signal generation process, and a band synthesis process of synthesizing the waveform signals of a plurality of bands subjected to the gain control correction process. In the method, the gain control correction processing is performed for each band in the waveform signal generation processing, and the gain control correction information includes flag information indicating whether to perform gain control, and performs actual gain control. The gain control correction information of the no block includes only the flag.

An information decoding apparatus according to the present invention comprises: decoding means for decoding a frequency component signal and gain control correction information; and a plurality of bands which are blocked from the frequency component signal at predetermined unit times. Waveform signal generation processing means for generating a waveform signal of the above, gain control correction processing means for performing gain control correction processing of an output waveform signal from the waveform signal generation processing means, and waveforms of a plurality of bands subjected to the gain control correction processing An information decoding apparatus having a band synthesis process for synthesizing a signal, wherein the gain control correction process is performed for each band in the waveform signal generation process, and the gain control correction information indicates whether to perform gain control. Wherein the gain control correction information of a block that does not actually perform gain control includes only the flag. .

Further, in the information recording medium according to the present invention, frequency component signal information and gain control correction information obtained by frequency component decomposition processing performed by blocking a signal on the time axis for each predetermined unit time are recorded. The gain control correction information includes flag information indicating whether or not to perform gain control. The gain control correction information of a block that does not actually perform gain control includes only the flag, Is selected.

[0041]

In these inventions, the frequency component decomposition processing includes a spectrum conversion processing for converting a signal on the time axis into a signal on the frequency axis. Also,
The input signal or the output signal may be an acoustic signal.

[0043]

[Function] By using a gain control amount selectively determined from a plurality of magnitudes as a gain control amount in a portion where a waveform signal sharply increases, gain control according to a degree of amplitude change of an attack portion can be performed. Thus, encoding, decoding, recording, and transmission with higher efficiency and higher sound quality can be performed.

[0044]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block circuit diagram showing an embodiment of an encoding apparatus to which the information encoding method of the present invention is applied. In FIG. 1, an audio signal input to an encoding device via an input terminal 100 is supplied to a band division circuit 1.
01 is divided into bands. This band division circuit 101
As the band dividing means in, a dividing means using a filter such as the above-described QMF may be used, or a means for grouping spectra obtained by spectrum conversion such as MDCT for each band may be used. Alternatively, a means may be used in which spectrum conversion is performed once on a band that has been divided into several bands by a filter, and the resulting spectrum is grouped for each band. Further, the width of each band by this band division may be uniform or may be non-uniform so as to match, for example, a critical bandwidth. In the example of FIG. 1, the frequency band is divided into four bands.
Alternatively, it may be reduced.

The signals divided by the band dividing circuit 101 are normalized by normalizing circuits 111, 112, 113, and 114 corresponding to each band for each certain time block. Decomposed into a normalized signal. The respective normalized signals are respectively quantized circuits 121, 122, 123, and 124 based on the quantization precision information output from the quantization precision determination circuit 141.
, And converted into a normalized / quantized signal. In FIG. 1, each of the quantization circuits 121, 122, 12 from the quantization precision determination circuit 141 is referred to.
Among the quantization accuracy information to 3, 124, the quantization accuracy information sent to the quantization circuit 122 is transmitted via a terminal 152,
The quantization accuracy information sent to the quantization circuit 123 is sent via a terminal 153, and the quantization accuracy information sent to the quantization circuit 124 is sent to a corresponding circuit via a terminal 154.

The quantization circuits 121, 122, 123,
The respective normalized / quantized signals from 124, the respective normalization coefficients from the above-mentioned normalization circuits 111, 112, 113 and 114, and the respective pieces of quantization accuracy information from the above-mentioned quantization accuracy determination circuit 141 are multiplexed. The code string is sequentially formed by 131, and this code string is output from the terminal 103. This code string is then recorded on a recording medium such as a disk, a tape, or a semiconductor, or transmitted from a transmission system.

Here, in the example of FIG. 1, the quantization precision determination circuit 141 calculates the quantization precision based on each signal divided by the band division circuit 101. It is also possible to calculate from the signal via the terminal 100 of
It is also possible to calculate based on the normalization coefficients from 1,112,113,114. Further, the calculation in the quantization accuracy determination circuit 141 can be performed based on an auditory phenomenon such as a masking effect, and the quantization accuracy information is output via the multiplexer 131 as described above. It is sent to the decoding device later. Therefore, the auditory model used in the encoding device can be set arbitrarily.

FIG. 2 is a block circuit diagram showing an embodiment of a decoding apparatus corresponding to the encoding apparatus shown in FIG. 1 to which the information decoding method of the present invention is applied. In FIG. 2, the code information (the code sequence) input to the terminal 201 of the decoding apparatus according to the present embodiment is sent to the demultiplexer 202, where the quantization accuracy information for each band and the normalization coefficient And a normalized / quantized signal. The quantization accuracy information, the normalization coefficient, and the normalized / quantized signal for each band are used as signal component configuration circuits 2 corresponding to the respective bands.
11, 212, 213, 214, where signal components are configured for each band. The signal components from the respective signal component constituting circuits 211, 212, 213, 214 are combined by the band combining circuit 221 to form an audio signal, which is output from the terminal 251.

Next, FIG. 3 is a diagram for explaining a gain control operation at the time of windowing processing when the embodiment of the present invention is applied.

Here, in the method described in the conventional example described above, the gain control amount for amplifying the portion immediately before the attack is fixed. For example, in JP-A-63-7026, each block is divided into sub-blocks,
The amplitude fluctuation between the two sub-blocks is a predetermined limit value (20
When the amplitude exceeds (about dB), a method is disclosed in which the amplitude of the signal before the amplitude changes abruptly is amplified by the encoding means and corrected by the decoding means.

However, for example, although the signal waveforms SW1 and SW2 in FIG. 3A both include an attack portion, there is a great difference in the manner in which the amplitude changes. That is, in the signal waveform SW1, there is a waveform signal of a certain level or more immediately before the attack portion, and the generated pre-echo is masked to some extent by the original waveform signal, though not as much as after the attack portion. On the other hand, in the signal waveform SW2, the level of the waveform signal immediately before the attack portion is very low, and the generated pre-echo is hardly masked.

Here, if the detection of the attack portion is identified by only one limit value of the level change and the same gain control and gain control correction are to be performed, the signal waveform SW
If the limit value and the gain control amount are set to be optimal for 1, the pre-echo will be heard for the signal waveform SW2. When the limit value and the gain control amount are set so as to be optimal for the signal waveform SW2, the signal waveform S2
Unnecessary gain control is performed on W1 and energy is spread in the frequency domain, so that the coding efficiency is reduced.

The method of the present invention solves this problem by changing the gain control amount according to the degree of amplitude change in the attack portion of the signal waveform.

That is, in the method of the present invention, as shown in FIG.
(B) performs gain control and gain control correction by applying the gain control function G1 having a relatively small gain control amount, whereas the gain control amount is relatively small for the signal waveform SW2. The gain control and the gain control correction are performed by applying the large gain control function G2.

FIG. 3C shows how the quantization noise is generated when the above processing is performed. As shown in FIG. 3C, the quantization noise before the attack portion of the quantization noise of the signal waveform SW1 has a relatively small noise suppression effect due to the gain control correction processing. Although the quantization noise is large compared to the quantization noise before the attack portion, the energy of the quantization noise is small throughout. On the other hand, the energy of the quantization noise throughout the signal waveform SW2 is relatively large, but the quantization noise before the attack portion is sufficiently low. Since the pre-echo is a great obstacle to hearing, it is desirable to suppress the noise by giving priority to the reduction of the overall noise energy.

Next, FIG. 4 shows an example of the flow of processing for detecting an attack portion and generating a gain control function when the embodiment of the present invention is actually applied to signal encoding. is there. For example, the encoding method of the present invention can be realized by incorporating this processing into the processing corresponding to the attack detection circuit 402 of the encoding apparatus in FIG.

In FIG. 4, for example, a block having a length of 2M is divided into N sub-blocks, and the maximum amplitude value P [I] in the I-th sub-block is divided into K consecutive blocks up to the I-th sub-block. Maximum amplitude value Q in sub-block
Compared to [I], if the ratio is equal to or more than a predetermined ratio, it is determined that an attack portion has been detected. Finally, a gain control function having a smooth transient portion is formed to prevent the spread of energy when converted into a spectrum.

That is, in the first step S1 in FIG. 4, I of the sub-blocks obtained by dividing one block into N
The maximum amplitude value Q [I] from the K consecutive sub-blocks up to the I-th sub-block, ie, the I-K + 1 sub-block to the I-th sub-block, is obtained.
The maximum amplitude value P [I] in the number-th sub-block is obtained. In the next step S3, I = 0, and in step S4
In the above, R as the above gain control amount is determined by the ratio of the maximum amplitude Q [I] of the K sub-blocks up to the I-th to the maximum amplitude P [I + 1] of the immediately following sub-block. T in the next step S5 is a predetermined threshold value,
Is larger than T, it is determined that an attack portion has been detected, and the process proceeds to step S9. If NO, the process proceeds to step S6, where I is incremented, and step S6 is performed.
At 7, it is determined whether or not I has reached the sub-block number N at the end of the block. Steps S4 and subsequent steps are repeated until I = N. When YES is determined in the step S7, L = 0 in step S8, that is, there is no attack, R = 1, and the process proceeds to the step S10. If YES in step S5, that is, if an attack is found, the process proceeds to step S9, where L = I + 1, and the integer of the value of R obtained in step S4 is substituted for R. That is, the length of the block before the attack portion is interpreted as L subblocks, and the value of R at this time represents the gain control amount. After finishing the processing in step S9, the process proceeds to step S10.

In step S10, the gain control function of the sub-block up to the attack position L is set to R, the remainder is set to 1 and an interpolation process is performed so as to finally have a smooth transition portion. are doing. That is, in this step S10, the gain control function g (n) is constructed based on the values of L and R. In the sub-block immediately before the attack portion, the function value is smoothly interpolated. This is to prevent the energy distribution from being spread when converted to the frequency domain, thereby enabling efficient coding.

As described above, by changing the gain control amount of the attack section according to the signal level, there is an advantage that the pre-echo can be effectively prevented even when the compression ratio is high.

In this example, the gain control is amplified only immediately before the attack portion. However, as described above, this utilizes the effect of forward masking in particular. However, of course, it is also possible to perform gain control so that amplification is performed in the small amplitude part during attenuation.For example, when the block length of the spectrum conversion is extremely long and the forward masking effect cannot be sufficiently expected, The signal may be amplified in a small amplitude portion at the time of attenuation. Also, the number of attack units to be detected does not necessarily need to be one for one block.

If a gain control function that changes abruptly in a step-like manner is used, when converted into a spectrum, the energy is spread and the coding efficiency is reduced. Therefore, it is desirable that the control function has a shape that smoothly changes to some extent even in the attack portion. However, since the pre-echo will be heard if the section is not sufficiently short, the gain control function has a transient section of about 1 msec in consideration of human hearing, and a smooth section such as a sine wave shape within the section. It is desirable to make a change. In the case where an attack occurs at the head of the next block, by expanding the detection range of the attack part to the sub-block at the head of the next block, the gain control function can have a smooth transition part.

As described above, the method or the apparatus of the present invention can be applied to an apparatus for processing a signal obtained by converting an acoustic waveform into a digital signal. For example, the present invention can be applied to the case where the processing is performed by the above method. Further, it is of course possible to record and transmit the code thus obtained on a recording medium. Further, the present invention can be applied to a case where encoding is always performed at a constant bit rate and a case where encoding is performed at a bit rate that changes with time so that the number of allocated bits differs for each block. It is.

In the above description, a case has been described where the waveform signal digitized by the encoding device is directly converted into a spectrum signal by using spectrum conversion. The method of the present invention can also be applied to a case in which a spectrum signal is converted using a spectrum conversion for each band.

FIG. 5 shows an example of a recording format when recording information encoded by the method of the present invention on a recording medium or a transmission format when transmitting the information.

In the example shown in FIG. 5, the code of each block is the same as the code of each block, depending on the contents of the attack part detection flag and the spectrum signal code, and the contents of the attack part detection flag. It is composed of information and gain control correction function generation information composed of gain control information. For example, the value of L in FIG. 4 may be recorded as the attack position information, and the value of R in FIG. 4 may be recorded as the gain control amount information. Since the proportion of blocks in which the attack portion where pre-echo is a problem exists in the actual music signal is low, it is efficient to record the attack position information and gain control amount information only in the block where the attack portion actually exists. Is good. However, it is needless to say that the gain control correction function generation information may be recorded in all blocks. In this case, in a block in which no attack portion actually exists, for example, L = 0 and R = 1 may be recorded. I just need.

Next, FIG. 6 shows an example of processing in which the decoding means generates a gain control correction function h (n) from the recording information shown in FIG.

For example, the processing shown in FIG.
The gain control correction function h (n) is generated by incorporating it into a process corresponding to the gain control correction circuit 414 of the first decoding device.
Is multiplied by the waveform signal element formed by the inverse spectrum conversion circuit 413, whereby the decoding method according to the present invention can be realized. Of course, the process of actually multiplying h (n) may be omitted in a block in which no attack portion is detected.

In the example of FIG. 6, step S21
To detect the attack detection flag, and when the flag is 0, that is, when no attack is detected, the process proceeds to step S22, where the gain control correction function h (n) is set to 1
And exit. If the flag is 1, that is, if an attack has been detected, the process proceeds to step S23, where the gain control function g (n) for L subblocks from the beginning of this block is set to R, and the above interpolation process is performed. The gain control function g (n) is obtained. In the next step S24, the reciprocal 1 / g (n) of the gain control function g (n) is calculated to obtain the gain control correction function h (n).

The method of the present invention can of course be applied to, for example, the method described in JP-A-3-132228 described above.

In addition to the case where a waveform signal is directly decomposed into frequency components by spectrum conversion, for example, the case where a waveform signal once band-divided by a band division filter is frequency-component decomposed by spectrum conversion, the present invention is of course also applicable. The method of the invention can be applied. Further, the present invention can be applied to a case where a waveform signal is decomposed into frequency components by a filter. The frequency components referred to in the present invention include all those obtained by these processes. However, when the pre-echo is applied in relation to the frequency components obtained by the processes including the spectrum conversion, which is a particularly serious problem, the present invention The method is particularly effective.

Further, the method of the present invention can be applied to an apparatus for processing a signal obtained by converting an acoustic waveform into a digital signal, and a waveform signal once stored in a file is processed by a computer or the like. It can also be applied to cases. Further, it is of course possible to record and transmit the code thus obtained on a recording medium. Further, the method of the present invention is applicable to a case where encoding is always performed at a constant bit rate and a case where encoding is performed at a bit rate that changes over time so that the number of allocated bits differs for each block. Is possible.

As described above, the case where the quantization noise when the acoustic waveform signal is quantized is made inconspicuous has been described. However, the method of the present invention makes the generation of quantization noise of other types of signals inconspicuous. The above is also effective, and can be applied to, for example, an image signal. However, since the pre-echo in the attack part of the sound signal causes a great obstacle to hearing, it is very effective to apply the present invention to the sound signal. Also, the method of the present invention is of course applicable to multi-channel audio signals.

[0075]

As is clear from the above description, the information encoding method according to the present invention is not limited to the part where the waveform signal sharply increases in the gain control processing of the input waveform signal to the frequency component decomposition processing. The gain control amount of the gain control process is selectively determined from a plurality of types of magnitudes, so that gain control can be performed in accordance with the degree of amplitude change in the attack portion, and more efficient and more sound quality can be obtained. Higher encoding is possible.

In the information decoding method according to the present invention, when performing the gain control correction processing of the output waveform signal in the waveform signal synthesis processing, the gain control of the gain control correction processing in a portion where the waveform signal suddenly increases is performed. By using a correction amount selected from a plurality of sizes determined based on the content of the gain control correction information, efficient processing can be performed and a high-quality signal can be reproduced.

These effects can be similarly applied to the information encoding device and the information decoding device.

By applying this to an audio signal or an audio signal, it is possible to prevent the occurrence of a pre-echo together with efficient processing.

Further, by recording or transmitting a signal encoded by such an encoding method or apparatus on a recording medium, efficient recording or transmission is possible.

[Brief description of the drawings]

FIG. 1 is a block circuit diagram illustrating a schematic configuration of an encoding device to which an embodiment of the present invention is applied.

FIG. 2 is a block circuit diagram showing a schematic configuration of a decoding device to which an embodiment of the present invention is applied.

FIG. 3 is a diagram for explaining an operation of gain control during windowing processing according to the embodiment of the present invention.

FIG. 4 is a flowchart schematically illustrating an example of a processing procedure for generating a gain control function in the encoding method according to the embodiment of the present invention.

FIG. 5 is a diagram showing a recording state of a code string obtained by encoding according to the embodiment of the present invention.

FIG. 6 is a flowchart schematically illustrating an example of a part of a processing procedure of a decoding method according to an embodiment of the present invention.

FIG. 7 is a diagram for explaining the operation principle of pre-echo generation in transform coding.

FIG. 8 is a diagram for explaining the operation principle of a conventional encoding / decoding technique using a variable conversion window length.

FIG. 9 is a diagram for explaining the operation principle of encoding and decoding using the conventional windowing processing technique.

FIG. 10 is a block diagram illustrating a schematic configuration of an encoding device based on a conventional windowing processing technique.

FIG. 11 is a block diagram showing a schematic configuration of a decoding device based on a conventional windowing processing technique.

[Explanation of symbols]

 Reference Signs List 101 band dividing means 111 to 114 normalizing circuit 121 to 124 quantizing circuit 131 multiplexer 141 quantization precision determining circuit 202 demultiplexer 211 to 214 signal component configuration circuit 221 band synthesizing circuit

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-3-263925 (JP, A) JP-A-5-304479 (JP, A) JP-A-4-304029 (JP, A) JP-A-5-304 91061 (JP, A) Special table Hei 6-507296 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H03M 7/30 G10L 19/02 G11B 20/10 301

Claims (5)

(57) [Claims] 1. A and band division processing for dividing the input waveform signal into a plurality of bands, the band and the gain control process for controlling the gain of the divided waveform signal, the gain control process is predetermined unit time input waveform signal
An information encoding method for performing frequency component decomposition processing of dividing the frequency component into frequency components and encoding processing of output information of the frequency component decomposition processing and control information of the gain control, wherein the gain control processing is performed. What is have lines for each band which is the band dividing process, the control information of the gain control, performing gain control
Do not perform actual gain control.
The control information of the new block consists of only the flag.
An information encoding method characterized by being performed. 2. A band division processing means for dividing an input waveform signal into a plurality of bands; a gain control processing means for performing gain control processing of the band-divided waveform signal; and an output signal from the gain control processing means. Predetermined unit time
Information encoding apparatus comprising: a frequency component resolution processing means for decomposing the frequency components into blocks, and encoding means for performing encoding of the control information of the output information and the gain control from the frequency component resolution processing means for each a is, if the gain control process, have rows for each band which is the band dividing process, the control information of the gain control, performing gain control
Do not perform actual gain control.
The control information of the new block consists of only the flag.
Information encoding apparatus which is characterized in that. 3. A decoding frequency component signal and a gain control correction information processing blocks for each predetermined unit time from the frequency component signal
Waveform signal generation processing for generating the obtained waveform signals of a plurality of bands, gain control correction processing of the output waveform signal of the waveform signal generation processing, and a band for synthesizing the waveform signals of the plurality of bands subjected to the gain control correction processing An information decoding method for performing synthesis processing, wherein the gain control correction processing is performed for each band in the waveform signal generation processing , and the gain control correction information indicates whether to perform gain control.
Includes flag information indicating the actual gain control is not performed
Is the gain control correction information of the block only the flag?
Information decoding wherein the al constructed. 4. A decoding means for decoding processing frequency component signal and a gain control correction information, blocks every predetermined unit time from the frequency component signal
Waveform signal generation processing means for generating waveform signals of a plurality of bands obtained, gain control correction processing means for performing gain control correction processing of an output waveform signal from the waveform signal generation processing means, An information decoding apparatus having a band synthesis process for synthesizing waveform signals of a plurality of bands, wherein the gain control correction process is performed for each band in the waveform signal generation process , and the gain control correction information is a gain control signal. Whether to do
Includes flag information indicating the actual gain control is not performed
Is the gain control correction information of the block only the flag?
Information decoding apparatus characterized by et constituted. 5. A signal on a time axis is blocked every predetermined unit time.
The frequency component signal information and the gain control correction information obtained by the frequency component decomposition processing performed by the conversion are recorded, and the gain control correction information indicates whether or not to perform the gain control.
Flag information that does not perform actual gain control.
The lock gain control correction information from the lock only
With the configuration, the information recording medium which is characterized in that which has been selected for each band of the frequency component signal.