JPH07106977A - Information decoding device - Google Patents

Abstract

PURPOSE:To reduce the noises due to the rasping pre-echoes by connecting together the waveform signals existing before the position shown by the dividing position information after inverting those signals in the opposite direction in terms of time. CONSTITUTION:The dividing position information given from a dividing position deciding circuit 502 is supplied to a terminal 602. The spectral signal supplied from a terminal 601 undergoes the inverse spectral transformation through an inverse spectral transformation circuit 603 corresponding to the spectrum transformation of a forward spectrum transformation circuit 503. Then the inversely transformed spectral signal is stored in a waveform signal storage circuit 604 and also sent to a waveform connecting circuit 605. The circuit 605 replaces the output of the circuit 603 with the signal obtained by processing the waveform signal of the precedent block stored in the circuit 604 on a time base based on the dividing position information received from the terminal 602 and the connects them together. The signals thus connected together by the circuit 605 are free from pre-echoes and taken out through a signal terminal 606.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information decoding apparatus in which input digital data is coded by so-called high efficiency coding, and what is transmitted, recorded and reproduced is decoded to obtain a reproduced signal. It is a thing.

[0002]

2. Description of the Related Art Conventionally, there are various techniques for highly efficient coding of a signal such as audio or voice. For example, an audio signal on a time axis is divided into blocks in a predetermined unit time, and each block is on the frequency axis. Blocking frequency band division method that transforms into a signal (spectral conversion) and divides into multiple frequency bands and encodes each band, that is, so-called transform coding, blocks audio signals on the time axis for each unit time There is band division coding (sub-band coding: SBC), which is a non-blocking frequency band division scheme that divides into a plurality of frequency bands and encodes without conversion. Further, a method of high efficiency coding in which the above band division coding and transform coding are combined is also considered, and in this case, for example, after performing band division by the band division coding, A signal for each band (a signal for each band on the time axis) is divided into blocks, a spectrum is converted to a signal on the frequency axis for each block, and coding is performed for each spectrum-converted band.

Here, as a band division filter used in the band division encoding of the above-mentioned high efficiency encoding, there is a filter such as a so-called QMF. QMF
1976 REC Crochiere Digital coding of speech in
Subbands Bell Syst.Tech.J. Vol.55, No.8 1976. Also, ICASSP 83, BOSTON PolyphaseQua
drature filters-A new subband coding technique Jos
eph H. Rothweiler describes a filter partitioning method with equal bandwidth.

As the spectral transform used in the transform coding, for example, the input audio signal is divided into blocks in a predetermined unit time (frame), and the discrete Fourier transform (DFT), cosine transform (DCT), There is spectrum conversion in which the time axis is converted into the frequency axis by performing modified DCT conversion (MDCT) or the like. Regarding the above MDCT, ICAS
SP 1987 Using FilterBank Designs Based on Time Do
main Aliasing Cancellation JPPrincen ABBradley
Univ. Of Surrey Royal Melbourne Inst. Of Tech.

According to the high-efficiency coding described above, the band in which the quantization noise is generated can be controlled by quantizing the signal divided into each band by the filter or spectrum conversion as described above, and masking can be performed. By utilizing properties such as effects, it is possible to perform auditory and more efficient encoding. Further, if the normalization is performed for each band, for example, by the maximum absolute value of the signal component in the band before the quantization is performed here, more efficient encoding can be performed.

Here, as the frequency division width for quantizing each frequency component obtained by frequency band division, there is, for example, band division in consideration of human auditory characteristics. That is, an audio signal may be divided into a plurality of bands (for example, 25 bands) with a bandwidth that is generally called a critical band and has a wider bandwidth in a higher band.
Also, when encoding the data for each band at this time,
Coding is performed by predetermined bit allocation for each band or adaptive bit allocation (bit allocation) for each band. For example, when the coefficient data obtained by the MDCT processing is encoded by the bit allocation, adaptive allocation bits are assigned to the MDCT coefficient data for each band obtained by the MDCT processing for each block. The encoding will be performed by numbers. The following two methods are known as bit allocation methods.

For example, IEEE Transactions of Accoustic
s, Speech, and Signal Processing, vol. ASSP-25, No.4,
In August 1977, bit allocation is performed based on the signal size of each band. In this method, the quantization noise spectrum becomes flat and the noise energy becomes the minimum, but the actual noise feeling is not optimal because the masking effect is not used auditorily.

Further, for example, ICASSP 1980 The critical b
and coder --digital encoding of the perceptual requ
irements of the auditory system MAKransner MIT
Describes a method of performing a fixed bit allocation by obtaining a necessary signal-to-noise ratio for each band 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.

In order to solve these problems, all bits that can be used for bit allocation are fixed bit allocation patterns that are predetermined for each small block that further subdivides the block, and the signal size of each block. The bit allocation is dependent on the input signal, and the division ratio depends on the signal related to the input signal. The smoother the spectrum of the signal, the larger the division ratio to the fixed bit allocation pattern. An efficient coding device has been proposed.

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, so that the overall signal-to-noise characteristic is significantly improved. Can be improved. In general, human hearing is extremely sensitive to a signal having a steep spectral component, so improving the signal-to-noise characteristic by using such a method not only improves the numerical value in measurement. It is effective for improving the sound quality in the sense of hearing.

Many other methods have been proposed for the bit allocation method, and if the model relating to hearing is further refined and the performance of the coding apparatus is improved, more efficient coding is perceived auditorily. Will be possible.

[0012]

By the way, when a method of decomposing a signal into frequency components and quantizing and coding the frequency components is used as described above, the frequency components are decoded and synthesized. Quantization noise occurs in the generated waveform signal.

Here, if the original signal component is abruptly changed, a portion where the original signal waveform is not large (for example, the signal waveform is not necessarily large before or after the portion where the abrupt change is made). However, the quantization noise on the waveform signal of (not necessarily) may become large. At this time, since the quantization noise generated in the portion where the original signal waveform is not large is not hidden by the simultaneous masking by the signal in the abruptly changing portion, the quantization noise becomes an auditory obstacle. For example, the quantization noise generated in this way at the attack portion where the sound becomes loud suddenly is called a pre-echo.

In particular, when the signal is decomposed into a large number of frequency components by using spectrum conversion, the time resolution becomes poor and the pre-echo occurs over a long period. Here, for example, if the conversion length of the spectrum conversion is shortened, the generation period of the above-mentioned quantization noise is also shortened, but if so, the frequency resolution is deteriorated and the coding efficiency in the quasi-stationary part is deteriorated.

The masking effect includes, in addition to the simultaneous masking effect, a forward masking in which a sound generated earlier in time conceals a sound generated later and a sound generated later in time. There is a retrograde masking for concealing a generated sound, but the retrograde masking is effective for a very short time as compared with the forward masking. Therefore, in particular, the quantization noise that occurs before the time when the sound abruptly increases (when viewed from the quantization noise, the above-mentioned sound that abruptly increases becomes a time later) is perceived as a pre-echo. It becomes a big obstacle to the above.

As a means for solving such a problem, a method has been proposed in which the conversion length is shortened at the expense of the frequency resolution only in the portion where the signal waveform changes abruptly. When this method is used, sufficient frequency resolution is ensured in the quasi-stationary portion, and the pre-echo in the attack portion also has a sufficiently short generation period, so that it is hidden by so-called retrograde masking. Good coding is possible.

However, in such a method in which the conversion length is variable, it is necessary to provide conversion means corresponding to conversion of different lengths in the encoding device and the decoding 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, for example, for each critical bandwidth. When encoding is attempted collectively, the number of spectra included in each critical band also differs, and the encoding and decoding processes become complicated.

As described above, there is a drawback that the encoding device and the decoding device are complicated by the conventionally known method of varying the conversion length.

Here, a state of generation of pre-echo when spectrum conversion is used in band division will be described below with reference to FIG.
It will be described using 0.

In FIG. 10, when quantization noise is added to the spectrum signal obtained by subjecting the waveform signal SW to forward spectrum conversion using the conversion window function as shown in FIG. Waveform signal S on the time axis again after inverse spectrum conversion is applied to the spectrum signal to which the quantization noise is added.
Returning to W, the quantization noise QN spreads over the entire transform block, as shown in FIG. That is, for example, in the case where the signal waveform SW suddenly increases in the middle of the conversion block as shown in FIG. 10B, a section in which the original signal waveform is small (for example, the signal rapidly increases as described above). In the section (before), since the quantization noise QN becomes larger than the original small signal waveform SW, simultaneous masking does not work, and it becomes a perceptual obstacle as a pre-echo.

Conventionally, in order to reduce the hearing loss due to such a pre-echo, the conversion length is shortened as described above, for example, as shown in FIG. Here, in general, for a quasi-stationary signal waveform, a longer conversion block length (using a long conversion window function having a longer conversion length as shown in (A) of FIG. 10) causes a specific spectrum. Although energy is concentrated on the coefficient, the coding efficiency is increased, but the pre-echo becomes a problem when the conversion block length is long in a portion where the loudness of the sound changes abruptly.

Therefore, when the above-mentioned signal suddenly becomes large, as shown in FIG. 11A, the conversion block length is shortened to perform spectrum conversion (conversion by the short conversion window function). As a result, the generation period of the pre-echo is shortened sufficiently (the generation period of the quantization noise QN is shortened). By doing so, so-called retrograde masking by the original signal comes to be effective, and the audible obstacle is eliminated. Conventionally, the conversion block length is selectively switched according to the property of each part of the signal waveform by using the method shown in FIG.

As a configuration for selecting the conversion block length as shown in FIG. 11, the configuration as shown in FIG. 12 can be taken as an example. The configuration of FIG. 12 is incorporated in the band dividing means in the case where the signal is divided into a plurality of bands in the encoding device and then the conversion is performed.

In FIG. 12, a waveform signal corresponding to the input signal (acoustic signal) to the encoding device is supplied to the input terminal 301, and this waveform signal is converted into the conversion length determining circuit 30.
Sent to 1. The conversion length determination circuit 301 checks whether or not the size of the waveform signal suddenly increases, and outputs a conversion block length control signal accordingly. This conversion block length control signal is sent to the forward spectrum conversion circuit 302, and is also sent to a decoding device or the like as a part of the output signal via a multiplexer or the like (not shown) via the terminal 304.

In the forward spectrum conversion circuit 302, the conversion block length based on the conversion block length control signal is
Spectral conversion such as DCT as described above is applied to the waveform signal from the terminal 301. The spectrum-converted signal is sent to a decoding device or the like via a terminal 303 as an output signal of each band through a multiplexer or the like (not shown).

The configuration on the decoding device side corresponding to the configuration of FIG. 12 is as shown in FIG. That is,
The configuration of FIG. 13 is incorporated in the band synthesizing means on the decoding device side corresponding to the band dividing means in the coding device.

In FIG. 13, for example, the conversion block length control signal is supplied to the terminal 401, the spectrum-converted signal is supplied to the terminal 402, and these signals are sent to the inverse spectrum conversion circuit 403. The inverse spectrum conversion circuit 403 inversely converts the spectrum-converted signal based on the conversion block length control signal. This inversely converted signal is taken out from the terminal 404.

The above-mentioned conventional method is capable of acoustically efficient coding for both the quasi-stationary part and the transient part of the signal waveform, but the conversion length is not constant. The number of spectra to be obtained also differs for each transform block, and there is a drawback in that both the encoding device and the decoding device become complicated.

Therefore, according to the present invention, not only the encoding device but also the decoding device can be simplified by making the conversion block length constant at the time of encoding and decoding, and the encoding in the quasi-stationary part is possible. It is an object of the present invention to provide an information decoding device that can improve the decoding efficiency and that can prevent the hearing loss due to pre-echo with a simple configuration.

As another method for solving the problem of pre-echo, the applicant of the present invention has previously filed Japanese Patent Application No. 5-18.
In the specification and drawings of No. 3988, the encoder is provided with means for detecting a position where a pre-echo is expected to be generated, the signal is temporally divided at this position, and each signal has a long conversion length. We have proposed a method that transforms into spectral signals, encodes them, and combines them on the time axis by a decoding device. By using this method, it is possible to prevent the occurrence of pre-echo while keeping the conversion length constant.

However, in the above-mentioned method of dividing the signal into front and rear in the section where the pre-echo occurs to obtain two spectrum signals, it is necessary to encode and record or transmit the two spectrum signals. It was difficult to implement if the number of bits available for optimization was very low.

Further, it is necessary to perform the spectrum forward conversion or the spectrum inverse conversion twice in each of the encoding device and the decoding device, which causes a problem that the amount of arithmetic processing increases.

[0033]

The present invention has been made in view of the above circumstances, and a first information decoding apparatus of the present invention decodes a quantized frequency component to obtain a waveform signal. A waveform signal generating means for obtaining a waveform signal from a quantized frequency component; a waveform signal processing means for storing the waveform signal and performing a predetermined processing on the waveform signal in a predetermined section; And a waveform replacement unit that determines a specific section of the signal and replaces the waveform signal of the specific section with the waveform signal of the section other than the specific section added by the waveform signal processing unit. .

Here, the information decoding apparatus of the present invention is carried out as follows. The frequency component is a signal obtained by spectrum conversion. Further, the conversion length of the spectrum conversion is fixedly set.

The waveform replacing means connects the waveform signals processed by the waveform signal processing means, which have been replaced as the waveform signals in the specific section, to obtain a continuous output waveform signal.
Further, the waveform signal processing means performs processing for inverting the waveform signal in the predetermined section, and the waveform replacing means replaces the inverted waveform signal as the waveform signal in the specific section to perform the connection, thereby performing the continuous operation. A typical output waveform signal. Similarly, the waveform signal processing means performs processing for repeatedly inverting the waveform signal in the predetermined section, and the waveform replacing means uses the repeatedly inverted waveform signal from the waveform signal processing means as the waveform signal in the specific section. Replace and perform the above connection. Similarly, the waveform signal processing means repeats processing of a section surrounded by samples having a value of substantially 0 as the predetermined section, and the waveform replacing means is a sample having a value of substantially 0 from the waveform signal processing means. The enclosed section is replaced by replacing the repeated waveform signal with the waveform signal of the specific section to perform the connection.

Further, the device of the present invention has division position information input means to which division position information indicating a specific section of the waveform signal is supplied, and the waveform replacement means has division position information from the division position information input means. Based on the above, the specific section is determined. As the waveform signal of the specific section, a section in which the waveform signal of the section other than the specific section is processed by the waveform signal processing unit is output, and the division position information is included in the unit section for obtaining the frequency component. It is the part before the position determined by.

Further, in the present invention, a plurality of frequency components are input to the same section, and other frequency components except at least one frequency component of the plurality of frequency components are used at the time of decoding. . The output signal after decoding is an acoustic signal.

The second information decoding apparatus of the present invention is
Inverse spectrum conversion means for performing inverse spectrum conversion on the quantized spectral component for each band, and storing a waveform signal of a specific band among the waveform signals of the respective bands subjected to the inverse spectrum conversion, and a waveform of a predetermined section of the specific band. Waveform signal processing means for subjecting the signal to predetermined processing, and a specific section of the waveform signal of the specific band among the waveform signals of the respective bands subjected to the inverse spectrum conversion, and a waveform signal of the specific section of the specific band. Of the waveform signals in the other band of the specific band excluding the specific band, the waveform replacing unit processing the waveform signal processed by the waveform signal processing unit, and the waveform signal of each band subjected to the inverse spectrum conversion. A waveform suppression unit for determining a specific section of the band other than the waveform signal of the specific band and suppressing the waveform signal of the specific section of the other band, and the waveform replacement unit. Those having a band synthesizing means for the other bands of waveform signals excluding the specific band from the constant band waveform signal and the waveform suppressing means band synthesis to obtain an output waveform signal.

Here, this information decoding apparatus is also configured as follows. For example, the above-mentioned spectrum component is obtained by a conversion length fixedly set.

The waveform replacing means connects the waveform signals processed by the waveform signal processing means, which are replaced as the waveform signals in the specific section of the specific band, to form a continuous output waveform signal. For example, the waveform signal processing means performs processing to invert a waveform signal in a predetermined section of the specific band, and the waveform replacement means performs the connection by using the inverted waveform signal as a waveform signal in the specific section of the specific band. Thus, the continuous output waveform signal is obtained. Similarly, the waveform signal processing means performs processing to repeatedly invert the waveform signal in a predetermined section of the specific band, and the waveform replacement means applies the repeatedly inverted waveform signal from the waveform signal processing means to the specific band. The connection is performed as the waveform signal of the specific section. Similarly, the waveform signal processing means performs processing for repeating a section surrounded by samples having a value of substantially 0 as the predetermined section of the specific band, and the waveform replacing means outputs the value of substantially 0 from the waveform signal processing means. The connection is performed by using the waveform signal in which the section surrounded by the samples is repeated as the waveform signal in the specific section of the specific band.

Further, the second information decoding apparatus of the present invention has division position information input means to which division position information indicating the specific section of the waveform signal is supplied, and the waveform replacement means and the waveform suppression. The means determines the specific section based on the division position information from the division position information input means. As the waveform signal of the specific section of the specific band,
The section for outputting the waveform signal obtained by processing the waveform signal of the other section of the specific band excluding the specific section by the waveform signal processing unit is more than the position determined by the division position information in the unit section for obtaining the frequency component. This is the previous part in time.

Also in the second information decoding apparatus of the present invention, a plurality of frequency components are input in the same band for the same section, and at least one frequency component of the plurality of frequency components is excluded. The frequency component is used at the time of decoding. The output signal after decoding is an acoustic signal.

[0043]

According to the present invention, for example, the position of the attack portion detected on the encoding side is supplied as division position information together with the encoded signal, and the waveform before the position indicated by the division position information is supplied to the information decoding device. Pre-echo is prevented by replacing the signal with a signal obtained by subjecting the previous section (block) to a predetermined process and synthesizing it with the waveform signals of the preceding and subsequent sections.

[0044]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described below with reference to the drawings.

First, prior to the description of the information decoding apparatus of the present invention, an information encoding apparatus corresponding to the information decoding apparatus of the embodiment of the present invention will be described with reference to FIG.

In FIG. 1, the audio signal input to the encoding device through the input terminal 100 is band-divided by the band-dividing circuit 101. As the band dividing means in the band dividing circuit 101, the above-mentioned QMF is used.
If a dividing means such as a filter is used,
A means of grouping the spectra obtained by the spectrum conversion of T etc. for each band may be used. Alternatively, a means may be used in which the spectrum is once converted into a plurality of bands divided by a filter, and the spectrum obtained by this is grouped for each band. Further, the width of each band resulting from this band division may be uniform, or may be non-uniform so as to match the critical bandwidth, for example. In the example of FIG. 1, the band is divided into four bands, but of course, this number may be further increased or decreased.

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

The quantizing circuits 121, 122, 123,
Each of the normalized / quantized signals from 124, the normalization coefficients from the normalization circuits 111, 112, 113, 114, and the quantization precision information from the quantization precision determination circuit 141 is a multiplexer. The (Multiplexer) 131 sequentially forms a code string, and this code string is output from the terminal 103. This code string is then recorded on a recording medium such as a disk or 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 band-divided by the band division circuit 101, but before the band division. It is also possible to calculate from the signal through the terminal 100 of each normalizing circuit 11
It is also possible to calculate based on the normalization coefficient from 1,112,113,114. Further, the calculation in the quantization precision determination circuit 141 can be performed based on an auditory phenomenon such as a masking effect, and each of the quantization precision information is output via the multiplexer 131 as described above. It is later sent to the decoding device.
Therefore, the auditory model used in the decoding device can be arbitrarily set.

On the other hand, FIG. 2 shows a block circuit diagram of an information decoding apparatus of an embodiment of the present invention corresponding to the encoding apparatus of FIG.

In FIG. 2, the code information (the code string) input to the terminal 201 of the decoding apparatus of this embodiment is
It is sent to the demultiplexer 202, where it is separated and restored into the quantization precision information for each band, the normalization coefficient, and the signal to be normalized / quantized. The quantization accuracy information for each band, the normalization coefficient, and the normalized / quantized signal are signal component configuration circuits 211, 212, 213, 21 corresponding to the respective bands.
4 and the signal components are formed for each band.
Each of these signal component configuration circuits 211, 212, 213, 2
The signal components from 14 are synthesized by the band synthesis circuit 221 into an audio signal and output from the terminal 251.

Next, the principle of the method for preventing the occurrence of the pre-echo in the above information encoding and decoding will be described with reference to FIG.

3A and 3B show the signal waveform SW.
And the state of generation of quantization noise QN when processing is performed with a fixed conversion block length similar to the conventional one (conversion by the window function for forward conversion shown in FIG. 3A). When processing is performed with a fixed conversion block length similar to this conventional one, FIG.
As shown in (B) of, the actual output signal after decoding is
The waveform signal SW and the quantization noise QN are added together.

Therefore, at the time of decoding in this embodiment,
As shown in (D) of FIG. 3, the output signal waveform SW of the section (specific section) shown by SE2 in the figure in which a pre-echo occurs with respect to the output signal waveform SW of (B) of FIG. Signal waveform S of the section shown (other predetermined section except the specific section)
I will replace it with the processed W.

That is, in this embodiment, the section SE2 is set.
As the output signal waveform SW of, the signal waveform SW of the section SE1 which is inverted in the temporally opposite direction (inverted in the temporally opposite direction) is used. Here, since the section SE2 is a short section of about several tens of msec at most, if the frequency component of the section SE2 is not significantly different from the frequency component of the original signal, the sense of discomfort in hearing will be small. By inverting and connecting the signal waveforms in the opposite direction in time, the signal waveforms SW of the section SE1 and the section SE2 are continuously connected, and annoying noise (noise due to pre-echo) is generated in the reproduced sound. There is nothing to do.

Of course, there are various methods other than the above for continuously connecting the signal waveforms in a certain section as described above.
It may be repeated once. Further, for example, it is also possible to adopt a method of cutting out a section sandwiched between two points (samples) in which the magnitude of the waveform signal is a value of almost 0 and repeating this section as many times as necessary. . In this case, if the interval is set to be an integral multiple of the fundamental wavelength of the audio signal, good approximation can be made to the original audio signal, which is advantageous in sound quality. In these cases,
The predetermined section is in the latter half of the section SE1 or between the two-point samples, and the folded and inverted sections of these sections or the sections repeated the necessary number of times are replaced with the specific section (SE2).

Here, in order to specify the length of the section SE2, the waveform signal is scanned by the encoding device and, for example, for each of the first to eighth small sections as shown in FIG. 3C. It is possible to take a method of examining the maximum absolute value of and identifying the small section in which the absolute value suddenly increased. This Figure 3
In the above example, the first to fourth small sections are specified as the pre-echo generation section.

That is, the flow of processing for obtaining the length N of the pre-echo generation section is as shown in the flowchart of FIG.

In the present embodiment, when the maximum absolute value of the sample in each small section is K times the maximum absolute value of the sample in the previous small section (K is a given value),
The length N of the pre-echo generation section is calculated assuming that the pre-echo occurs in the sub-section before the sub-section.

In FIG. 4, first, the value of the length N is initialized to 0 in step S1, and in the next step S2, for example, the maximum absolute value of the sample of the first small section of FIG. Let P ₀ . In step S3, the variable I indicating the number of the small section is set to 1, and in step S4, the absolute value of the sample of the next small section, that is, the (I + 1) th small section is set to P ₁ .

In step S5, the absolute value P ₁ of the sample in the (I + 1) th small section is compared with K times the maximum absolute value P ₀ of the sample in the previous section, and P is compared. _{When 1} > K * P ₀ (Yes), the process proceeds to step S7. When P ₁ > K * P ₀ does not hold (No), the process proceeds to step S6.

In step S6, the value of I is set to the length N, and then the process ends.

In step S7, the above P ₀ and P ₁ are compared. If P ₁ > P _{0 in} step S7 (Yes), the process proceeds to step S8, and if P ₁ > P ₀ is not satisfied (No), the process proceeds to step S9.

In step S8, P ₀ = P ₁ is set, and in step S9, it is determined whether or not the variable i is 7, that is, whether or not the variable has reached the 8th small section. In this step S9, i = 7
If it is determined to be Yes (Yes), the process is terminated, and if No is determined, the process proceeds to step S10.

In step S10, after setting I = I + 1,
Returning to step S4, the above-mentioned processing is repeated.

From the encoding device to the decoding device, the length N of the pre-echo generation section obtained as described above is the division position information (signal switching information) for preventing pre-echo.
Sent as.

That is, the section where the pre-echo that occurs before the attack portion where the signal changes abruptly is the specific section SE2, and the length N from this attack section to the position where the pre-echo does not occur is divided position. Information. By sending this division position information to the decoding device, as will be described later, the decoding device can replace the section SE2 indicated by this division position information with the processed section SE1.

Of course, such a method of detecting the attack portion includes the case where the length of the small section is set to 1 (that is, in sample units). That is, the division position information (signal switching information) is not the number of small sections,
It is also possible to use the number of samples up to the time of division (time of switching). However, if the pre-echo generation period is sufficiently short, it is possible to prevent the above-mentioned retrograde masking from disturbing the auditory sense. Therefore, recording information in a small number of sections can improve the coding efficiency.

Next, FIG. 5 shows a specific configuration of the band division circuit of the encoding apparatus of this embodiment having a configuration for determining the division position.

In FIG. 5 showing a specific example of the band division circuit 101 of FIG. 1, the waveform signal corresponding to the input signal from the input terminal 100 of FIG. 1 is sent to the division position determination circuit 502 and the forward spectrum conversion circuit 503. . The division position determination circuit 502 detects the position where the signal sharply increases, for example, by the procedure shown in the flowchart of FIG.
This is output from the terminal 504 as division position information. The detection signal from the terminal 504, that is, the division position information, is omitted in FIG.
As a part of the output signal from the output terminal 103 through
It is sent to a decoding device or recorded on a recording medium.

In the forward spectrum conversion circuit 503, the waveform signal from the terminal 501 is subjected to spectrum conversion such as DCT as described above with a fixed conversion block length. This spectrum-converted signal is sent to the subsequent stage configuration via the terminal 505. That is, as a part of the output signal from the output terminal 103 through the multiplexer 131 of FIG. 1, it is sent to the decoding device or recorded on the recording medium.

The configuration of the decoding apparatus side of the embodiment of the present invention corresponding to the configuration of FIG. 5 is as shown in FIG. That is, the configuration of FIG. 6 is incorporated in the band synthesizing circuit 221 on the decoding device side of FIG.

That is, in FIG. 6, each of the signal component forming circuits 211 to 2 of the decoding apparatus of FIG. 2 is connected to a terminal 601.
A spectral signal from 14 is provided. Also, terminal 6
Although not shown in FIG. 2, the division position information (that corresponds to the output of the terminal 504 in FIG. 5) is supplied to 02. The spectrum signal from the terminal 601 is subjected to inverse spectrum conversion processing corresponding to the spectrum conversion in the forward spectrum conversion circuit 503 of FIG. 5 by the inverse conversion spectrum circuit 603, and then the waveform signal storage circuit 60.
4 is stored and sent to the waveform connection circuit 605.

The waveform connection circuit 605 has the terminal 60
The output of the inverse spectrum conversion circuit 603 and the waveform signal of the previous block (section) stored in the waveform signal storage circuit 604 are processed based on the division position information from 2 (the signal waveform is temporally reversed). (Reversed in the direction)
Replace (switch) on the time axis and connect as described above. That is, the waveform signal storage circuit 604 is installed to temporarily store the waveform signal of the previous section.

The signal connected by the waveform connection circuit 605 has no pre-echo as described above, and this signal is taken out from the terminal 606.

Next, FIG. 7 shows the configuration of a band division circuit of another embodiment.

In another embodiment shown in FIG. 7, the input signal is divided into two bands in advance by the band division filter 702 before the forward spectrum conversion is performed. The terminal 701 of FIG. 7 corresponds to the terminal 501 of FIG. 5, the terminal 708 of FIG. 7 corresponds to the terminal 504 of FIG. 5, and the terminals 706 and 707 of FIG. It corresponds to. The division position determination circuit 705 in FIG. 7 also corresponds to the division position determination circuit 502 in FIG.

The forward spectrum conversion circuits 704 and 703 of FIG. 7 perform forward spectrum conversion on one of the corresponding outputs divided by the band division filter 702. In the example of FIG. 7, the low-frequency side output divided by the band division filter 702 is sent to the forward spectrum conversion circuit 703, and the high-frequency side output is sent to the forward spectrum conversion circuit 704. The outputs of these forward spectrum conversion circuits 703 and 704 respectively correspond to terminals 7.
It is output from 06, 707.

In the configuration of FIG. 7, since the band division filter 702 divides the band in advance to thin out the signal samples, the block length in calculating the forward spectrum conversion is maintained at the frequency resolution. It can be shortened as it is.

The structure of the band synthesizing circuit of the decoding device corresponding to the band dividing circuit of the other embodiment shown in FIG. 7 is as shown in FIG.

In FIG. 8, terminals 801, 802,
Signals corresponding to terminals 706, 707, and 708 of FIG. 7 are supplied to 803, respectively. That is, the signal supplied through the terminal 801 (the signal on the low frequency side) is sent to the inverse spectrum conversion circuit 804, where the inverse spectrum conversion corresponding to the forward spectrum conversion in the forward spectrum conversion circuit 703 of FIG. 7 is performed. After being processed, it is sent to the waveform connection circuit 807. The signal supplied through the terminal 802 (the signal on the high frequency side) is sent to the inverse spectrum conversion circuit 805, where the inverse spectrum conversion corresponding to the forward spectrum conversion in the forward spectrum conversion circuit 704 of FIG. 7 is performed. After being processed, it is sent to the waveform suppression circuit 808. These waveform connection circuits 807
The division position information from the terminal 803 is also supplied to the waveform suppression circuit 808.

The output of the inverse spectrum conversion circuit 804 is stored in the waveform signal storage circuit 806, and the output of the waveform signal storage circuit 806 is sent to the waveform connection circuit 807. ing. That is, in the waveform connection circuit 807, the output (the signal in the specific section) of the inverse spectrum conversion circuit 804 corresponding to the low frequency side (specific band) and the previous block stored in the waveform signal storage circuit 806. A waveform signal on the low frequency side (a signal of a section other than the specific section of the specific band) that has been processed is replaced (switched) on the time axis based on the division position information and connected.

Here, the characteristic of the configuration of FIG. 8 is that
It is the point that only the specific band (only the low band side in the example of FIG. 8) is used that is obtained by processing the previous waveform signal in the section immediately before the waveform suddenly increases. In the other band (high band), the signal obtained by the inverse spectrum conversion from the inverse spectrum conversion circuit 805 in the specific section of the high band is obtained by the waveform suppression circuit 808 based on the division position information. It is forcibly suppressed. As described above, the use of the processed waveform signal of the previous waveform signal only in the specific band (low band side) is performed by the waveform signal storage circuit 806.
It is convenient to reduce the storage capacity of. As this specific band (low band), it is preferable to set it to the low band side where the signal energy is usually dominant, as in the present embodiment.

The signal connected by the waveform connection circuit 807 has no pre-echo as described above, and the outputs of the waveform connection circuit 807 and the waveform suppression circuit 808 are sent to the band synthesis filter 809. The band synthesis filter 809 performs band synthesis corresponding to the division by the band division filter 702 of FIG. The band-combined signal is taken out from the terminal 801.

Furthermore, as one of the great advantages of the decoding method in the decoding apparatus according to the present invention, the information previously proposed by the applicant of the present application in the specification and drawings of Japanese Patent Application No. 5-183988. Encoding method and compatibility with a recording medium can be mentioned. In the method proposed in the previously proposed specification, etc., instead of processing the waveform signal of the previous block as in this embodiment and using it, the spectrum conversion is separately performed in the section immediately before the waveform suddenly increases, The spectral information is also sent to the decoding device to prevent the auditory disturbance due to the pre-echo.

That is, the above-mentioned information coding method proposed above is an information coding method in which a waveform signal is decomposed into spectral components, and the spectral components are quantized and coded. When the signal switching information is generated, decomposed into the spectral components, quantized and coded, the signal switching information is coded and the spectral components are calculated multiple times in the same section for a predetermined signal. It is characterized by that.

In the information decoding method of the embodiment of the present invention, the signal switching information in the previously proposed information encoding method can be used as the division position information in the present invention.

A more specific description will be given. Recording of the code encoded by the above-described information encoding method on the recording medium can be represented as shown in FIG. 9 when compatible with the present invention.

In FIG. 9, as the block information (only the arbitrary Nth block information is shown in FIG. 9), the spectrum information after switching (that is, the spectrum signal code SD1 before the processing in the section SE1) is recorded first. ,
Next, division position information (signal switching information) is recorded, and finally spectrum information before switching (that is, the section S
The spectrum signal code E2 of E2 is recorded.
As the division position information, a method of recording the number of small sections in FIG. 3 before switching can be used.

If the division position information is 0, it means that the switching and combining of the signal waveforms is unnecessary. In that case, the recording of the spectrum information before the switching can be omitted. At this time, it is possible to combine the signal from two spectrum signals of the same block by using the information decoding method described in the above-mentioned specification or the like, and the information decoding apparatus of the present invention. It is also possible to disregard the second spectrum signal (spectrum signal code SD2) of FIG.
In the information decoding apparatus of the present invention, it is sufficient to perform the inverse spectrum conversion once in the same block, so that the amount of calculation can be reduced.

As described above, according to the information decoding apparatus of the present invention, especially the applicant of the present invention first applies in Japanese Patent Application No. 5-18398.
Pre-echo can be prevented by a simple method while having compatibility with the information encoding or information decoding method and the recording medium proposed in the specification and drawings of No. 8.

In the decoding by the decoding apparatus according to the embodiment of the present invention, the separation of the waveform signal does not necessarily have to be performed in order to deal with the case where the signal waveform becomes abruptly large. May be performed in order to cope with the case where the value becomes suddenly small. This way,
It is possible to reduce the quantization noise in the portion where the signal becomes small. Further, the number of waveform signals generated by separating one signal waveform is not necessarily two, and may be more. However, since the biggest problem in hearing is when the signal waveform suddenly becomes large, it is very effective to use the decoding as in the decoding device of the present invention to prevent pre-echo. Further, in that case, since the waveform signals of the previous section are combined, the circuit can be simple and convenient.

Although the MDCT has been taken as an example of the spectrum conversion as described above, it goes without saying that the discrete Fourier transform (DF
It is also possible to use T) or discrete cosine transform (DCT). Further, the coding method of the present invention can be applied to the case where the band is divided by the filter without using such a special spectrum conversion and then the quantization is performed. In the description of the present invention, the frequency component thus obtained by the filter is also called spectrum information.

The method of the present invention can of course be applied to multi-channel acoustic signals.

Although the description has been made above regarding the case where the quantization noise when the acoustic waveform signal is quantized is made inconspicuous, the decoding by the information decoding apparatus of the present invention is the quantization of other kinds of signals. It is also effective in making noise generation inconspicuous, and can be applied to image signals, for example.

[0096]

As is apparent from the above description, by using the information decoding apparatus of the present invention, pre-echo can be prevented without changing the conversion section length, and not only the decoding apparatus but also the coding apparatus can be used. The configuration of the device can be simplified, and since it is not necessary to perform spectral conversion or inverse conversion in duplicate in each section, it is not necessary to increase the calculation amount of the encoding device and the decoding device so much, and recording Alternatively, it is not necessary to increase the spectrum information to be transmitted, so that the compression rate at the time of encoding can be increased.

[Brief description of drawings]

FIG. 1 is a block circuit diagram showing a schematic configuration of an encoding device according to an embodiment.

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

FIG. 3 is a diagram for explaining the operation principle of pre-echo suppression.

FIG. 4 is a flowchart showing a flow of processing for obtaining a length of a pre-echo generation section.

FIG. 5 is a block circuit diagram showing a specific configuration of a main part of a band division circuit of the encoding apparatus according to the present embodiment.

FIG. 6 is a block circuit diagram showing a specific configuration of a main part of a band synthesizing circuit of a decoding device according to an embodiment of the present invention.

FIG. 7 is a block circuit diagram showing a specific configuration of a main part of a band division circuit of an encoding device according to another embodiment.

FIG. 8 is a block circuit diagram showing a specific configuration of a main part of a band synthesizing circuit of a decoding device according to another embodiment of the present invention.

FIG. 9 is a diagram for explaining a method of recording an encoded signal.

FIG. 10 is a diagram for explaining pre-echo.

FIG. 11 is a diagram for explaining the operation principle of the conversion length variable and the effect thereof in the prior art.

FIG. 12 is a block circuit diagram showing a specific example of a main part of a band division circuit according to a conventional technique.

FIG. 13 is a block circuit diagram showing a specific example of a main part of a band synthesis circuit according to a conventional technique.

[Explanation of symbols]

 101 ... Band division circuit 111-114 ... Normalization circuit 121-124 ... Quantization circuit 131 ... Multiplexer 141 ... Quantization accuracy determination circuit 202. Multiplexers 211 to 214 ...- Signal component configuration circuit 221 ...- Band synthesis circuit 502,705 ...- Division position determination circuit 503,703,704 ... Forward spectrum conversion circuit 603,804,805 ... Inverse spectrum conversion circuit 604, 806 ... Waveform signal storage circuit 605, 807 ... Waveform connection circuit 702 ... Band division filter 808 ... Waveform suppression circuit 809 ... Band Synthesis filter

Claims (21)

[Claims] 1. A waveform signal generating means for obtaining a waveform signal from a quantized frequency component, a waveform signal processing means for storing the waveform signal and subjecting the waveform signal in a predetermined section to predetermined processing, and Waveform determining means for determining a specific section and replacing the waveform signal of the specific section with the waveform signal processed by the waveform signal processing means of the waveform signal of the other section except the specific section. Information decoding device. 2. The information decoding apparatus according to claim 1, wherein the frequency component is a signal obtained by spectrum conversion. 3. The information decoding apparatus according to claim 2, wherein the conversion length of the spectrum conversion is fixedly set. 4. The waveform replacing means connects the waveform signals processed by the waveform signal processing means, which have been replaced as the waveform signals of the specific section, to form a continuous output waveform signal. The information decoding device according to Item 1, 2 or 3. 5. The waveform signal processing means performs processing for inverting the waveform signal in the predetermined section, and the waveform replacing means replaces the inverted waveform signal as the waveform signal in the specific section to perform the connection. ,
The information decoding apparatus according to claim 4, wherein the continuous output waveform signal is obtained. 6. The waveform signal processing means performs processing for repeatedly inverting the waveform signal in the predetermined section, and the waveform replacing means applies the repeatedly inverted waveform signal from the waveform signal processing means to the waveform in the specific section. The information decoding apparatus according to claim 4, wherein the continuous output waveform signal is obtained by replacing the signal as a signal and performing the connection. 7. The waveform signal processing means performs processing for repeating a section surrounded by samples having a value of substantially 0 as the predetermined section, and the waveform replacement means has the value of substantially 0 from the waveform signal processing means. 5. The continuous output waveform signal is obtained by replacing the repeated waveform signal in the section surrounded by the samples taking the above with the waveform signal of the specific section and making the connection. Information decoding device. 8. A division position information input means to which division position information indicating a specific section of the waveform signal is supplied, and the waveform replacement means is based on the division position information from the division position information input means. The information decoding apparatus according to claim 1, 2, 3, 4, 5, 6, or 7, wherein a specific section is determined. 9. A section for outputting a waveform signal obtained by processing the waveform signal of the section other than the specific section by the waveform signal processing means as the waveform signal of the specific section is a unit section for obtaining the frequency component. 9. The information decoding apparatus according to claim 8, wherein the information decoding apparatus is a portion that is temporally preceding the position determined by the division position information. 10. A plurality of frequency components are input to the same section, and other frequency components except at least one frequency component of the plurality of frequency components are used at the time of decoding. Claim 1 characterized in that
The information decoding device according to 2, 3, 4, 5, 6, 7, 8, or 9. 11. The output waveform signal after decoding is an acoustic signal, as claimed in claim 1.
The information decoding device according to 6, 7, 8, 9, or 10. 12. An inverse spectrum conversion means for performing inverse spectrum conversion on a quantized spectral component for each band, and storing a waveform signal of a specific band among the waveform signals of each band subjected to the inverse spectrum conversion, and storing the specific band. Waveform signal processing means for performing a predetermined process on the waveform signal in the predetermined section, and a specific section of the waveform signal in the specific band among the waveform signals in the respective bands subjected to the inverse spectrum conversion is determined, and Waveform replacing means for replacing the waveform signal of the specific section with the waveform signal processed by the waveform signal processing means of the waveform signal of the other section of the specific band except the specific section; A waveform suppressing unit that determines a specific section of the band other than the waveform signal of the specific band in the waveform signal and suppresses the waveform signal of the specific section of the other band; Information having a band synthesizing unit for band synthesizing a waveform signal of a specific band from the shape replacing unit and a waveform signal of a band other than the specific band from the waveform suppressing unit to obtain an output waveform signal. Decoding device. 13. The information decoding apparatus according to claim 12, wherein the spectral component is obtained by a conversion length fixedly set. 14. The waveform replacing means connects the waveform signals processed by the waveform signal processing means replaced as the waveform signals in the specific section of the specific band to form a continuous output waveform signal. The information decoding device according to claim 12 or 13, characterized in that. 15. The waveform signal processing means performs processing for inverting a waveform signal in a predetermined section of the specific band, and the waveform replacing means replaces the inverted waveform signal as a waveform signal in the specific section of the specific band. 15. The information decoding apparatus according to claim 14, wherein the continuous output waveform signal is obtained by performing the connection by the above-mentioned connection. 16. The waveform signal processing means performs processing for repeatedly inverting the waveform signal in a predetermined section of the specific band, and the waveform replacing means specifies the repeatedly inverted waveform signal from the waveform signal processing means. By replacing the waveform signal of the specific section of the band and making the above connection,
15. The information decoding apparatus according to claim 14, wherein the continuous output waveform signal is obtained. 17. The waveform signal processing means repeats processing of a section surrounded by samples having a value of substantially 0 as a predetermined section of the specific band, and the waveform replacing means performs the processing of repeating the section from the waveform signal processing means. It is possible to obtain the continuous output waveform signal by replacing the repeated waveform signal in the section surrounded by the samples having a value of 0 with the waveform signal in the specific section of the specific band and performing the connection. 15. The information decoding device according to claim 14, wherein the information decoding device is a device. 18. A division position information input unit to which division position information indicating the specific section of the waveform signal is supplied, wherein the waveform replacement unit and the waveform suppression unit are divided from the division position information input unit. 14. The specific section is determined based on position information, as claimed in claim 12,
The information decoding device according to 14, 15, 16 or 17. 19. The waveform signal of the specific section of the specific band, which is obtained by processing the waveform signal of the other section of the specific band except the specific section by the waveform signal processing means, is output as the waveform signal of the specific section. 19. The information decoding apparatus according to claim 18, wherein the unit section for obtaining the frequency component is a portion temporally preceding the position determined by the division position information. 20. Inputting a plurality of frequency components in the same band for the same section, and using other frequency components except at least one frequency component of the plurality of frequency components at the time of decoding. 20. The information decoding device according to claim 12, 13, 14, 15, 16, 17, 18, or 19. 21. The output signal after decoding is an acoustic signal.
The information decoding device according to 16, 17, 18, 19, or 20.