JPH01194531A - Encode method for digital audio data - Google Patents

Abstract

PURPOSE:To improve error correcting ability by transmitting a change part as well concerning the data of a high important degree when digital audio data are data-compressed and transmitted. CONSTITUTION:A transmitting system by an ADPCM to data-compress the digital audio data and to transmit the data is composed of an encoder 10, a signal transmitting system 30 and a decoder 40. Here, concerning the forecasting parameter of the high important degree, these parameters are recorded and the difference between the forecasting parameters to be correspondent between two blocks is obtained in every two blocks and this difference is recorded to the area of a remainded word. Thus, error correction is executed without fail to the forecasting parameter of the high important degree and the error correcting ability can be improved.

Description

[Detailed description of the invention] The explanation will be given in the following order.

A Industrial field of application B Overview of the invention C Prior art D Problem to be solved by the invention E Means for solving the problem (Fig. 1) F Effect G Embodiment G1 First embodiment (Fig. 1) ~FIG. 4) G2 Other Embodiments H Effects of the Invention A Industrial Application Field This invention relates to a method for encoding digital audio data.

B. Summary of the Invention This invention enables better decoding by transmitting changes in highly important data when compressing and transmitting digital audio data. be.

C. Conventional technology For example, in 8 mm video, as an optional function, the audio signal is digitized into a PCM signal during recording, this PCM signal is recorded in the overscan section of the tape, and the reverse process is performed during playback. It is accepted that the original audio signal can be obtained by performing

In this case, if the sampling frequency and number of quantization bits of the PCM signal are increased, it is possible to record and play back the audio signal with better characteristics. It ends up.

Therefore, when recording, the number of bits of the PCM signal is compressed.
It has been considered to expand the number of bits during reproduction so that excellent recording and reproduction characteristics can be obtained even if the number of focuses on the tape is small.

AD is a method of bit compression/expansion.
There is a method called PCM.

FIG. 8 shows an example of a transmission system using ADPCM. In this example, each 64 consecutive samples of input data is treated as one block, and the prediction coefficients of the prediction filter are optimized for each block. This is the case when controlling the value. At this time, bit-compressed main data is output for each sample of the human data, and auxiliary data related to the bit compression is output for each block.

That is, in the same figure, (10) is the encoder, (3
0) is a signal transmission system, and (40) is a decoder. For example, if the PCM sound is used for 8mm video, the encoder (10) is installed in the recording string, and the decoder (40) is The transmission system (30) is provided on the recycled yarn and includes an error correction processing circuit, a nine-rotation magnetic head, and the like.

Then, in the encoder (10), the digital data Xt is transmitted in parallel for each sample from the input terminal (11) through the delay circuits (12) and (13) to the subtraction circuit (
14). In this case, the input data Xt is a linear A/D converted analog audio signal.
It is a CM signal, for example, the sampling frequency is 48k.
)lx, the number of quantization bits is 16 bits. Moreover, as shown in FIG. 9, the data Xt is -1≦xt<i
It is assumed that the data is expressed as a fixed-point number and as a two's complement number (the same applies to other data).
.

Furthermore, the delay circuits (12) and (13) are for synchronizing the timing of the main data and the auxiliary data, and each has a delay time of l block period (
Therefore, strictly speaking, if the input value of the terminal (11) is Xt, the output of the delay circuit (13) will be Xt-B a, but for the sake of complexity, it is written as Xt on the skewer).

Further, the predicted value Xt for the data Xt is taken out from the prediction filter (19), and the difference in this value is sent to the subtraction circuit (14).
The difference Dt between the value Xt and the value t is extracted from the subtraction circuit (14). This value Dt is the error (prediction residual) of the predicted value x [with respect to the input value Xt. Therefore, the value D
Ideally, t is Dt = 0, and it is generally a small value, so even if the word length of the value Dt is, for example, 16 bits, when Dt≧0, as shown in FIG. , all of the considerable bits on the MSH side become “0”, and when Dt<0, all become “1”, and the remaining few bits on the LSB side correspond to the difference between the value Xt and ×L. When the value Dt becomes a large value, the lower bits can be ignored.

Therefore, this value Dt is supplied to the gain control circuit (15) and multiplied by 0 (G≧1), resulting in a normalized value D
t-G, and this value G-IJt is sent to the requantization circuit (16
) and requantized, for example, into a 4-bit value ff1t-G.

Further, this value 5t-G is supplied to the gain control circuit (17) and multiplied by 1/G, thus giving a non-normalized value '6t of the same order as the value 1)t, and this value t5t is At the same time, the predicted value 5(t) from the filter (19) is supplied to the adder circuit (18), and from the adder circuit (18), the sum of the value et and the value t [ The value t-gt+et is taken out and this value t is fed to the filter (19).

In this case, the value t is the predicted value for the value Xt, and the value bt is the value obtained by truncating or rounding the 0-order bits of the error Dt at the time of prediction, so the value t and 15t are different. The value 55t which is the sum is almost equal to the input value Xt, and since this value t was supplied to the filter (19), the value 55t which is the filter output
can be a value that predicts the input value Xtφ1 at the next sampling point.

Then, the value t5t-G from the requantization circuit (16) is
The signal is supplied to a decoder (40) through a transmission system (30).

In this decoder (40), the value 15t-G is multiplied by 1/G by the gain control circuit (41) to give the value 5t, and this value at is supplied to the addition circuit (42), whose addition output is output from the output terminal. A prediction filter (43) extracted from (44) and configured similarly to filter (19)
and its filter output is supplied to the adder circuit (42).

Therefore, the output of the filter (43) becomes the value xt, and digital data t approximately equal to the input data Xt is taken out to the terminal (44).

Further, in order to set the prediction coefficients in the filters (19) and (43) to optimal values for each block, the following circuit is provided.

That is, the prediction filters (19) and (43) are
For example, it is a third-order filter that uses partial autocorrelation coefficients (PARCOR coefficients) as prediction coefficients, and its first to third order coefficients a1 to a3 can be changed to arbitrary values. .

Input data Xt from the terminal (11) is supplied to a time window circuit (21), subjected to predetermined weighting, and then supplied to an autocorrelation circuit (22) to calculate a correlation coefficient. The coefficients are supplied to a prediction coefficient circuit (23), and a partial autocorrelation coefficient of 1 to 3 is calculated as a prediction coefficient up to the third order for each block of data Xt, and a value of 1 to 3 is applied to this coefficient by a filter. (19) and the latch (
51) to the filter (43).

Further, data Xt from the delay circuit (12) is supplied to a prediction error filter (24), and the filter output is supplied to an intra-block maximum value detection circuit (25).

In this case, the filter (24) is the prediction filter (19)
a third-order prediction filter (241) configured in the same manner as
a subtraction circuit (242), and a coefficient circuit (242).
3) are supplied to the filter (241) as prediction coefficients from 1 to 3, and the predicted value (
5t (prediction error) is generated for each sample. Further, the detection circuit (25) detects the prediction error σt within the block for each block of input data Xt.
(There are 64 of these), the absolute value j5+ax of the prediction error with the largest absolute value is detected.

Then, this maximum value t5vaax is supplied to the normalized gain calculation circuit (26) and the normalized gain GG'' b /
t5max b is a safety factor of Q<b<1, and is converted to, for example, b-0,9, and this gain coefficient G for normalization is supplied to the gain control circuits (15) and (17), It is supplied to the gain control circuit (41) through the latch (52). In this case, the value j5max is the maximum value of the 64 values ff1t, so the value Dt-G is -1≦Dt・G<1
normalized to .

Note that Lunch (51) and (52) have coefficients of 1 to
3, to hold G for one block period of the corresponding value f5t-G.

Also, considering the amount of data transmitted from the encoder (10) to the decoder (40) via the transmission system (3o), the prediction residual t5t-G, which is the main data, is transmitted for each sample in 4 bits, for example. The prediction coefficients 1 to 3 and the gain coefficient G, which are auxiliary data (prediction parameters), are, for example, 16 bits, 12 bits, 12 bits, and 8 bits.
Bits are transmitted for each block, so the data in one block period is 4 bits x 64 samples + 16 bits + 12 bits +
It becomes 12 bits + 8 bits - 304 bits. and,
The amount of data for one block period when data compression is not performed is 16 bits x 64 samples - 1024 bits.

Therefore, the amount of data was compressed and transmitted to 304 bits/1024 bits = 29.7%.

In this way, according to this system, data compression of digital audio data can be performed, but in this case,
In particular, according to this system, even if there is a limit on the double length of the coefficients and calculations, the prediction coefficients of the prediction filters (19) and (43) are controlled to optimal values according to the input data Xt, so that the decoded Errors caused by compression of the data father t can be minimized.

In addition, when transmitting the prediction residual l)t, this residual Dt is reduced in number of bits by recombination, and normalization is performed before quantization, so that the transmitted data 6t -G is data with a small number of bits and little error.

FIGS. 5 and 6 show an example in which the encoder (10) and decoder (40) described above are used in an 8 mm video audio signal recording system and playback system.

That is, in the recording thread, for example, N 1' S C
The color video signal of the system is supplied to the recording video circuit (62) through the terminal (61), and the luminance signal is converted into an FM signal, and the carrier color signal is on the lower frequency side than the FM luminance signal, that is, ,til! The transmission frequency fc is f
c = 47.25fh ('-743kHz, f
h is the horizontal frequency), and these F
A sum signal Sv of the M luminance signal, the carrier color signal subjected to low frequency conversion, and the pilot signal for tracking servo during playback is extracted, and this signal Sv is sent to the recording amplifier (63).
It is supplied to the switch circuit (64) through.

In addition, the stereo left and right channel audio signals R and R are connected to the A/D converter (72L), (72)1) through the terminals (71L) and (71R).
Each of the signals H is A/D converted into digital data Lt and Rt with a sampling frequency of 48 kHz and a quantization bit number of 16 bits, and these data LL and RL are processed in the same way as the encoder (10) described above. Configured encoder (IOL).

(IOR) for each channel, the data f
it-G, kt~ki, data XLt corresponding to G;

L k i~Lka, LC; and XRt, Rkt~R
k3°RG is taken out.

These data are then supplied to the recording encoder (73), which performs processes such as adding error correction data, interleaving, and time axis compression to approximately 115 field periods at the end of each field period. The recording encoded digital signal Sa is supplied to a modulation circuit (74) and is converted into, for example, a pie-phase mark signal sb.This signal sb is passed through a recording amplifier (75) to a switch circuit (64). ).

Then, the switch circuit (64) is controlled at a predetermined timing so that fK'+SV is alternately supplied to the rotating magnetic heads (IA) and (IB) every one field period, and the signal sb is supplied to the signal qSv. The signals are supplied to the heads (IA) and (1B) in a reverse relationship.

Moreover, the heads (16) and (IB) are 18
has an angular spacing of 0゛, is rotated at a frame frequency in synchronization with the color video signal at the terminal (61), and
The magnetic tape (2) is run diagonally at a constant speed over an angular range of more than 216 degrees around its rotating circumferential surface. In addition,
The heads (LA) and (IB) have different slit angles, so-called azimuth angles.

Therefore, as shown in FIG.
Trunks (2'r) are successively formed adjacent to each other, and 1
The signal sb for the field period is recorded and remains 180
The signal Sv for one field period is recorded in the section .

Note that the recording system and recording format described above are the same as the current 8
It is similar to millivideo.

On the other hand, in the reproduction system, the head (1^) sequentially reproduces the signals Sv and Sb from every other track (2), and the head (IB) sequentially reproduces the signals Sv and Sb from every other track (2) remaining.
Signals S v + S b are sequentially reproduced from T), and these reciprocating signals are sent to a reciprocating amplifier (81A).

(81B) to the switch circuit (82), and from the switch circuit (82) head (1^), (IB
) are successively taken out, and the signals sb reproduced by the head (1^) and (IB) are
, sb are taken approximately every 115 field periods at the end of each field period.

Then, the signal Sv from the switch circuit (82) is supplied to the reproduction video circuit (83), where the reverse processing to that during recording is performed and the original color video signal is taken out to the terminal (84).

Also, the signal sb from the switch circuit (82) is ? The signal Sa is demodulated by being supplied to the M modulation circuit (91), and this signal Sa is supplied to the reproduction decoder (92) to perform time axis expansion, deinterleaving, error correction, etc. Data with XLt ~L
G, XRt-RG are decoded, and these data are sent to a decoder (4
0L), (40R) and the data Lt, R
t is decoded and these data are sent to the D/A converter (
93L), (93R) and terminal' (9
4L) and (94R), the original audio signal R is extracted.

As described above, the audio signals R, , R are recorded and reproduced. In this case, when calculating the amount of data in one field period, the amount of data of the prediction residuals XLt and XRt is as follows.
48000 samples/field frequency -48000/ per channel
59.94... = 800.8 samples, but since it is not possible to round down the fraction, it is 80.
1 sample = 801×4 bits. Therefore, for 2 channels of signal R, 801x 4 bits x 2 channels - 801 words (l
word - 8 bits).

In addition, the amount of data for the prediction coefficients LkL to Lk, Rkl to Rk3 and gain coefficients LG and RG is 64 samples (1
Since one set is obtained for each block), there are 801 samples/64 samples #12.51 sets per channel, but since we cannot round down the fraction, we have 13
For two channels, signal R and H, there are 26 sets. When this is converted into the number of words, it becomes (16+12+12+8 bits) x 26 sets = 6 words x 26 sets - 156 words.

Therefore, the total amount of data in one field period is 801 words + 156 words = 957 words.

On the other hand, when calculating the amount of digital audio data in the current 8mm video, the sampling frequency is 2rh'' = 31.468kHz and the number of star bits is 8 bits, so the amount of data in the field period is With 2 channels of R, 2 r h7 field frequency x 2 channels = 525
Sample x 2 channels - 1050 words.

Therefore, since the amount of data due to the above-mentioned ADPCM is smaller than the amount of PCM audio data in the current 8 mm video, the encoder (73) converts each data due to the above-mentioned ADPCM into the digital audio data in the current 8 mm video. The recording can be encoded by assuming that Also, this allows the encoder (73) - tape (2) -> decoder (9)
The signal system 2) is unchanged from the current 8 mm video, that is, it is possible to record and reproduce audio signals with better sound quality while using the current tape format.

Literature: "Fundamentals of Speech Information Processing", patent application published by Ohmsha, 1986
-299285 Specification and Drawing D Problems to be Solved by the Invention By the way, in the above-mentioned 8 mm video, even if an error occurs in the decoded signal Sa, the decoder (92)
In this case, if errors in data XLt-LG and XRt-RG can be completely corrected, no problem will occur.

However, in reality, when performing encoding processing for recording in the encoder (73), only one set of P parity, Q parity, and CRC code is added for each field period for error correction. Therefore, complete error correction may not be possible with this method.

And among the data XLt-LG, XRt-RGO), the coefficient L k L - L k 31 L G + Rk
If an error occurs in t~Rk3°RG, this is 1
Since this relates to the data Lt and Rt of the block period, it has a large effect on the auditory sense.

Therefore, it is necessary to add even stronger error correction capability to the coefficients Lk1 to LG and Rkt to RG.

However, according to the above numerical example, the data capacity of current 8 mm video in one field period is 1050 words, whereas the above ADPCM uses 957 words, so per field period , 1050 words - 957 words = only 93 words are left, and within these 93 words, the coefficients Lkt ~ LG, Rki
~The error correction ability for RG must be improved.

This invention attempts to answer such a need.

E Means for Solving the Problem Therefore, in the present invention, the highly important prediction parameters Lkz~LG and Rkt~RG are recorded as described above, and the two blocks are recorded for each two blocks. The difference in the corresponding prediction parameters between the two blocks is determined, and this difference is recorded in the remaining 93 word area.

F: Error correction for prediction parameters with high operational importance is reliably performed.

G Example G1 First Example In the case of the above numerical example, in one field period, the child series l
Parameters L k 1 to Lk3. Since the prediction parameters Rk1 to Rki for 13 blocks of RG and the prediction parameters Rk1 to Rki for 13 blocks of RG are obtained, in the following explanation, these prediction parameters are expressed as data P L n + P Rn (n
is a block number, collectively referred to as nc=1 to 13). That is, for example, the data PLz is the parameter L in the first block of the left channel in a certain field period.
ks~Lk3. LG is shown as a representative.

FIG. 1 shows an example of a recording system, in which the prediction residual XLt is supplied from the left channel encoder (IOL) to the interface circuit (101), and the encoder (101) is supplied to the left channel encoder (IOL).
As shown in FIG. 3B, data PLn is extracted from 0L) for each block period, and this data PLn
n is supplied to the interface circuit (101).

Furthermore, the data PLn from the encoder (IOL) is supplied to the subtraction circuit (102), and the delay circuit (102) is supplied to the subtraction circuit (102).
03) and is delayed by one block period as shown in C in the figure, and this data PLn-
z is supplied to the subtraction circuit (102), and as shown in the figure, the difference data 5LnSLn -PLn -PLn-t
” (i) However, for SLn, n-1~
Only odd numbers and n-12 of it are taken out as valid, and this data SLn is supplied to the interface circuit (101).

In this case, for example, if data PLn is the coefficient Lk1 of the second block and data PLn-1 is the coefficient Lk s of the first block, the data SLn is
This is the difference between these two coefficients. That is, data SLn is the difference between two corresponding prediction parameters in two consecutive blocks.

Further, similar processing is performed on the right channel data XRt, Rn, resulting in data XRt, Rn.

SRn (=PRn-PRn-t) is supplied to the interface circuit (101).

In the interface circuit (101),
Since the residuals XLt and XRt are each 4 vias per sample, the left channel residual XLt and the right channel residual t-XRt corresponding to the sampling time are one word that occupies the upper 4 bits and the lower 4 bits, respectively. It is combined into data Mi.

In this case, as mentioned above, in one field period, the remaining!
! XLt and XRt are obtained for 801 samples×2 channels, and therefore, 801 words (i-0 to 800) of data Mi are obtained during the I field period.

Furthermore, in the interface circuit (101),
Data PLn, PRn (parameters Lk1 to LG
, Rkx~RG) is reconfigured into data Pj (j=0~155) of 8 bit length x 156 words, and
Data SLn and SRn are (1B+ 12+ 12+
8 bits) x 7 sets x 2 channels = 6 words x 14 sets of data Ss (mmQ~83).

These data Mi, Pj + Ss are sent to the encoder (73) as PCM in current 8mm video.
The signal Sa is supplied in place of the audio data, and subjected to recording encoding processing, that is, addition of error correction data, interleaving, time axis compression, etc., every one field period, and is extracted as a signal Sa. After this signal Sa is converted to signal sb, the head (1^), (
IB) is recorded on tape (2).

In this case, data Mi: 801 words, data Pj: 156 words, data Sm: 84 words per field period, resulting in a total of 1041 words, which fits within the current 8 mm video data 1l 1050 words. and therefore can be recorded in the current formant.

FIG. 2 shows an example of recycled yarn, in which the left channel data Mi, Pj, Ss+ are decoded and taken out from the decoder (92), and these data are supplied to the interface circuit <201) to be processed as shown in FIG. 4A.

As shown in C and F, the original data X L t + P
The data XLt is decoded into L n +SLn, and is delayed by one block period as shown in FIG. 4B by the delay circuit (202) for timing adjustment with other data PLn and SLn, and then supplied to the decoder (40L). be done.

Also, data P from the interface circuit (201)
Ln is sequentially supplied to the delay circuits (203) and (204), and as shown in E in the figure, data PLn-t and PLn-2 delayed by one block period are sequentially taken out, and data PLn-1 is sent to the swissona circuit. (205) to the decoder (40L), and the data PL
n''-PLn-z is supplied to the arithmetic circuit (206).

Furthermore, the data SLn from the interface circuit (201) is supplied to the arithmetic circuit (206), and is also supplied to the delay circuit (207), where it is delayed by one block period as data 5Ln-x, as shown in FIG. , this data 5Ln-1 is supplied to the arithmetic circuit (206).

Then, in the arithmetic circuit (206), data PLn is obtained based on equation (i). That is, from equation (i), PLn-1=PLn-5Ln-(ii) or,
From formula (i), PLn = PLn-t + SLn,', PLn-1-PLn-z+5Ln-t'' (
iii).

Therefore, when the block number (n-1) in data PLn-1 (D in the same figure) is an even number, C1F in the same figure, ! -1
As shown by the broken line in , data PL is calculated based on equation (ii).
Data PLn-1 is calculated from n and SLn, and when the block number (n-1) is an odd number and (n-1) = 12, data PLn-1 is calculated based on formula (iii) as shown in E, G, and H of the same figure. The data PLn-s is calculated from the data PLn-2+S Ln-t, and these calculated data P
Ln-1 is supplied to the switch circuit (205).

Furthermore, the decoder (92) includes an error detection circuit (21).
1) is connected and an error has occurred in the data Pj, a detection signal Se indicating this is extracted, and this signal Se
is delayed for one block period by the delay circuit (212) for timing adjustment, and then supplied to the switch circuit (205) as a control signal, and in the event of an error, the switch circuit (205)
5) is switched to a state opposite to that shown in the figure.

According to such a configuration, during reproduction, data XLt is supplied to the encoder (40L), and data Pj
In other words, if no error occurs in the data P obtained from the delay circuit (203') after one block period,
When no error occurs in Ln-1, the switch circuit (205) is in the state shown in the figure, and data PLn-1 (D in the figure) from the delay circuit (203) is sent to the decoder (40L).
The data Lt is decoded.

However, if the data Pj has an error that cannot be corrected by the decoder (92), then the data Pj obtained from the delay circuit (203) after l block period
If an error occurs in Ln-1, the data P
j is taken out from the delay circuit (203) as data PLn-z, the signal Se causes the switch circuit (2
05) is switched in the opposite direction as shown in the figure, data PLn-1 from the arithmetic circuit (206) is switched to the decoder (40L).
) and the data Lt is decoded.

Similar processing is also performed for the right channel.

Thus, according to the present invention, when an error occurs in data PLn-1 due to an error in data Pj, (
Since the data PLn-s is extracted using the data of the adjacent block based on the formula (ii) or (iii), the error correction ability for the data PLn-* is improved, the signal can be properly reproduced, and the same The signal R can also be properly reproduced.

Moreover, in this case, for each two blocks, the corresponding prediction parameter L k 1 between the two blocks
~LG, Rk 1~RG difference data SLn.

SRn is obtained, and this difference data is used as the original data XLt.

Since they are recorded together with L k 1 to LG, XRt, and Rkt to RG, the difference data SLn and SRn can be stored in the remaining 93 words as described above.
There is no need to change the recording format, etc.

G2 Other Embodiments In the above description, difference data SLn is obtained from two blocks in each of the left and right channels.

SRn is obtained, but the left channel block data PL
It is also possible to obtain difference data between PRn and data PRn of the right channel block, and record and reproduce this data in the same manner as described above.

H Effects of the Invention According to this invention, data Pj is changed due to an error in data Pj.
When an error occurs in Ln-r, data PLn-1 is extracted using the data of the adjacent block based on equation (ii) or (iii), so the error correction ability for data PLn-s is signal R can be appropriately reproduced, and in the same way, signal R can also be appropriately reproduced.

Moreover, in this case, for each two blocks, the corresponding prediction parameter L k 1 between the two blocks
~LG, Rks~RG difference data SLn.

Obtain SRn and use this difference data as original data XLt.

Since they are recorded together with Lk1 to LG, XRt, and Rks to RG, the difference data SLn and SRn can be stored in the remaining 93 words as described above, and there is no need to change the recording format.

[Brief explanation of the drawing]

FIG. 1 is a system diagram of an example of the present invention, and FIGS. 2 to 9 are diagrams for explaining the same. (10) is an encoder, (30) is a signal transmission system, (40
) is a decoder.

Claims (1)

[Claims] A predetermined number of A/D converted samples are set as one block, a prediction parameter is obtained from the samples included in the block for each block, and based on this prediction parameter, A prediction residual is obtained for each sample of the block for which this prediction parameter was obtained, and this prediction residual and the above prediction parameter are sent out, and for each two of the blocks, the corresponding A method for encoding digital audio data, comprising: obtaining difference data between prediction parameters; and transmitting this difference data together with the prediction residual and the prediction parameter.