JPH01175326A - Method for encoding digital audio data - Google Patents

Abstract

PURPOSE:To attain a more excellent decoding by transmitting doubly a high- order bit of high significance out of data concerning the data of high significance when digital audio data are data-compressed and transmitted. CONSTITUTION:Data Dt.G of a predictive residual, predictive coefficients k1-k3, and gain data G are regarded as the digital audio data in a current 8mm video and an encoding is executed at every one field period. At this time, the highorder bit or high significance out of the data k1-k3 and G is assigned to residual 93 words and the high-order bit of high significance of the data k1-k3 and G is written doubly. When an error is generated in the data k1-k3 and G of high significance, the bit written doubly is substituted. Thus, even when an uncorrectable error is generated in the original predictive parameter, an audio signal without problems on auditory feeling can be reproduced. Even when there is a limitation on a recording capacity, an error correcting capacity to the predictive parameter of high significance is improved and the excellent decoding can be attained.

Description

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

A. Field of industrial application B. Outline of the invention C. Prior art D. Problem to be solved by the invention E. Means for solving the problem (Fig. 1) ~Figure 3) H Effect of the Invention A Field of Industrial Application This invention relates to a method for encoding digital audio data.

B. Summary of the Invention This invention provides a method for compressing and transmitting digital audio data by doubly transmitting the most important high-order bits of the data. This allows for an excellent decoat.

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

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 proposed to expand the number of bits during playback so that excellent recording and playback characteristics can be obtained even if the number of bits on the tape is small.

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

FIG. 4 shows an example of a transmission system using ADPCM. In this example, each 64 consecutive samples of human 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 l 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 it is applied to a PcM audio system in 8 mm video, the encoder (10) is provided on the recording thread, and the decoder (40) is provided on the playback thread. The transmission system (30) includes an error correction processing circuit, a one-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 XL is a linear A/D converted analog audio signal.
It is a CM signal, for example, the sampling frequency is 48k.
Hz, and the number of quantization bits is 16 bits. Furthermore, as shown in FIG. 5, it is assumed that the data Xt is expressed as a fixed point with -1≦Xt<1 and as a two's complement (the same applies to other data).

Further, 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 one block period (
Therefore, strictly speaking, if the human power value of the terminal (11) is Xt, the output of the delay circuit (13) will be Xt-128, but for the sake of complexity, it will simply be written as Xt).

Further, the predicted value ×t for the data Xt is taken out from the prediction filter (19), and this value ×L is sent to the subtraction circuit (14).
The difference between the values Xt and xt (1)t+Jt=xt-';<t is extracted from the subtraction circuit (14). This value D( is the error (prediction residual) of the predicted value nuL with respect to the human power value
Ideally, t is Di = 0, which is also a small value for boat fishing, so even if the word length of the value Dt is, for example, 16 bits, as shown in FIG. Sometimes, the significant bits on the MSB side are all “0”
When D(<0, all bits become "1", and the remaining few bits on the SB side become "O" or "1" corresponding to the difference between the value Xt and xL. Also, the value Dt
When 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-Dt is used in the requantization circuit (16)
and is requantized into, for example, a 4-bit value t5L-G.

Further, this value t5t-G is supplied to the gain control circuit (17) and multiplied by 1/G, thus resulting in an unnormalized value 5t of the same order as the value DL, and this value 1
5t is supplied to the adder circuit (18), and the predicted value t from the filter (19) is supplied to the adder circuit (18), which outputs the sum of the value t and ×t. The value t=nut+5t is taken out and this value t is fed to the filter (19).

In this case, the value statement t is the predicted value for the value The value that is the sum 9. t is approximately equal to the input value Xt. Since this value father t is supplied to the filter (19), the value 5<
t can be a value that predicts the input value Xt◆1 at the next sample time.

The value 6t-G from the rekeying circuit (16) is then supplied to the decoder (40) through the transmission system (30).

In this decoder (40), the value 6L-G is multiplied by 1/G by the gain control circuit (41) to obtain a value t5t, and this value 5t is supplied to the adder 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 the terminal (44) receives digital data 9. which is approximately equal to the input data Xt. t is taken out.

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
As a prediction coefficient, for example, partial autocorrelation coefficient (PAIICOR
It is a third-order filter that uses gA coefficients), and the coefficients a1 to a3 of the first to third orders of gA can be changed to arbitrary values.

Further, input data Xt from the terminal (11) is supplied to a time window circuit (21) to perform a predetermined vehicle finding, and then supplied to an autocorrelation circuit (22) to calculate a correlation coefficient. This coefficient is supplied to the 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 calculated for this coefficient. It is supplied to the filter (19) and the latch (
51) to the filter (43).

Furthermore, data Xt from delay circuit 1 (3) (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 t-measurement filter (241) configured in the same manner as
The subtraction circuit (242) and the coefficient circuit (242) are connected.
3) to the prediction coefficient from 1 to 1 (3 is the filter (241)
The predicted value (prediction error) t5t of the error let with respect to the input data Xt is generated every l samples. Further, the detection circuit (25) inputs the input data Xt
For each block, the absolute value j5max of the prediction error having the maximum absolute value among the prediction errors t5t (there are 64) in that block is detected.

Then, this maximum value j5max is calculated by the normalized gain calculation circuit (
26) and the data of the gain G during normalization, Q =
b/i5max b is a safety factor of 0<b<1, and is converted to, for example, b=o, 9, and this data G is used in the gain control circuit (15).

(17) and is also supplied to the gain control circuit @ (41) through the launch (52). In this case, the value 5
Since max is the maximum value of the 64 values f5t, the value Dt-G is normalized to -1≦Dt-G<1.

Note that latches (51) and (52) hold 1 for data.
to 3, G for one block period of the corresponding value 5t-G.

Also, considering the amount of data transmitted from the encoder (10) to the decoder (40) via the transmission system (30), data t>t-c, which is the main data, is transmitted for each sample using 4 bits, for example. The prediction coefficients □ to 3 and the gain data G, which are auxiliary data (prediction parameters), are transmitted for each block in, for example, 16 bits, 12 bits, 12 bits, and 8 bits, so the amount of data in one block period is , 4 bits x 64 samples + 16 bits + 12 bits +
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 transmitted compressed to 304 bits/1024 bits, or 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 to the word length of the coefficients and calculations, the prediction coefficients of the prediction filters (19) and (43) are controlled to optimal values according to the human data Xt, so the decoding Errors caused by compression of the data 5t can be minimized.

In addition, when transmitting the prediction residual l)t, this residual Dt is requantized to reduce the number of pips, and the transmitted data is normalized before the requantization. t5t·G has a small number of bits and is data with small errors.

! FIG. 86 and FIG. 7 show an example in which the encoder (10) and decoder (40) described above are used in a recording system and a reproducing system for 8 mm video audio signals.

3- That is, in the recording thread, for example, N 1' S
Then, a 'Ji color video signal is supplied to a recording video circuit (62) through a terminal (61) and a luminance signal or an FM signal is supplied to the recording video circuit (62).
At the same time as the carrier color signal is converted into a signal, the carrier color signal is on the lower frequency side than the FM luminance signal, that is, the first general transmission frequency fc is
f c -47,25f I+ (? 743kll
z, f and h are horizontal frequencies), and a sum signal Sv of these FM luminance signals, the low-frequency converted carrier color signal, and a pilot signal for tracking servo during playback is extracted. The signal Sv is supplied to the switch circuit (64) through the recording amplifier (63).

In addition, stereo left and right channel audio signals R and R are supplied to A/D converters (72L) and (72R) through terminals (71L) and (71R), respectively, and each R has a sampling frequency, for example. The data is A/D converted to 48kHz, 16-bit digital data xt, xt, and these data xt, xt are supplied to encoders (IOL) and (IOR) configured similarly to the encoder (10) described above. For each channel, data t5t・G,
3. G is taken out.

Then, these data are supplied to a recording encoder (73) and encoded for each field period, including addition of error correction data, interleaving, and time axis compression to approximately 115 field periods at the end of each field period. The processed digital signal Sa is supplied to a modulation circuit (74) and is converted into, for example, a pi-phase mark signal sb, and this signal sb is sent to a switch circuit (64) through a recording amplifier (75). Supplied.

Then, the switch circuit (64) is controlled at a predetermined timing so that the signal Sv is alternately supplied to the rotating magnetic heads (IA) and (1B) every one field period, and the signal sb is different from the signal Sv. In the reverse relationship, the head (LA
), (IB).

Also, the heads CIA) and (IB) are 18
It 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 just over 216° on its rotating circumferential surface. In addition,
The heads (LA) and (IB) have different slit angles, so-called azimuth angles.

Therefore, as shown in FIG.
Tracks (2T) are formed adjacently and sequentially,
1 in the 36° section from the beginning of the trunk (21')
The signal sb for the field period is recorded, and the remaining 180
In the section ``, the signal Sv for one field period is recorded.

Note that the recording system and recording format described above are the same as those of the current 8 mm video except for the signal Sa in the signal sb.

On the other hand, in the reproduction system, the head (IA) sequentially reproduces the signals Sν and Sb from every other track (2T), and the head (IB) sequentially reproduces the signals Sv and Sb from every other track (2T).
Signals Sv and Sb are sequentially reproduced from T), and these reproduced signals are sent to a reproduction amplifier (81^).

(81B'') to the switch circuit (82), and from the switch circuit (82) the head (IA),
The reproduced signals Sv, Sv of (IB) are taken out continuously, and the reproduced signals sb, sb of heads (IA), (IB) are taken out 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).

Further, the signal sb from the switch circuit (82) is supplied to the demodulation circuit (91) to demodulate the signal Sa, and this signal Sa is supplied to the reproduction decoder (92) for time axis expansion and
By performing deinterleaving and error correction, the original data f5t-c,k is restored for each channel.
t~ki, G are decoded, and these data are sent to a decoder (40L) configured similarly to the above-mentioned decoder (40).
), (40R) and the data 5<t, 5<
t is decoded, these data are supplied to 1)/A converters (93L) and (93R), and the original audio signal R is taken out at terminals (93L) and (94R).

As described above, the audio signal R is recorded and reproduced. In this case, when calculating the data loss that falls within one field period, the amount of data for the prediction residual 5t-c is 48,000 per channel. Sample/field frequency -48000/
50.94... = 800.8 samples, but I can't round down the fractions, so 80
1 sample = 801×4 bits. Therefore, for 2 channels of signal R and H, 801x 4 bits x 2 channels - 801 words (1
word - 8 bits).

Also, the amount of data for the prediction coefficients 1 to 3 and the gain data G is 1 set for every 64 samples (1 block), so for each channel, 801 samples/64
The sample is 12.51 pairs, but since we cannot round down the fraction, it becomes 13 pairs, and the signal is
In H channel 2, there are 26 pairs. Then, converting this 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 the field period is 801 words + 156 words = 957 words.

On the other hand, when calculating the amount of digital audio data in the current 8 mm video, the sampling frequency is 2 f h = 31.468 kHz, i
Since the l quantization bit number is 8 bits, the amount of data in the field period is 2 channels (signal and H), and is 2 [h/field frequency x 2 channels = 525 samples x 2 channels = 1050 words.

Therefore, since the amount of data by ADPCM mentioned above is smaller than the amount of PCM audio data in the current 8 mm video, the encoder (73) converts each data by the above ADPCM as it is into the digital audio data in the current 8 mm video. The recording can be encoded by assuming that Also, as a result, 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: [Basics of sound delivery information processing] Specification and drawing D of Japanese Patent Application No. 1983-299285 published by Ohmsha Problems to be solved by the invention By the way, in the above-mentioned 8 mm video, the demodulated signal Even if an error occurs in Sa, no problem will occur if the decoder (92) can completely correct the error in 3, G in data t5t-G, , kt~.

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.

Then, data [5t-G, kt~ is 3. As can be seen from the number of bits allocated to G, the prediction coefficient is 1
If an error occurs in ~3, especially if an error occurs in a low-order prediction coefficient, the accuracy of the decoded data will be greatly reduced, and the auditory sensation will be greatly affected.

Therefore, data 1 to 3. It is necessary to add even more powerful error correction capability to G.

However, according to the above numerical example, the data capacity of current 8mm video in one field period is 1050 words, whereas the above ADPCM uses 957 words, so one field period For each period, there are only 1050 words - 957 words - 93 words left over, and within these 93 words, the data is 1 to 3. The error correction ability for G must be improved.

This invention attempts to answer such a need.

E Means for Solving the Problems Therefore, in the present invention, for example, for each field period, all data t5t -G, kt~ are divided into 3
．． G is encoded by regarding it as digital audio data in the current 8mm video, and at this time, the remaining 93 words are assigned 1 to 1 to the most important data.
3. The higher-order bits of G with higher importance are assigned to data with higher importance of 1 to ka, and the higher-order bits of G with higher importance are written twice.

F When an error occurs in 1 to ki or G in data with high action importance, the double-written bit is substituted.

G Example G1 First Example In the example shown in FIG. 2, in the encoder (73), the first-order coefficient is set to 1 to 3 for the prediction coefficient and for the gain data G, as shown by diagonal lines. 1 (total 16 bits) upper 8 bits U
4 Upper 4 bits U of 2 (total 12 bits) for the second coefficient
2 Upper 4 bits U of 3 (total 12 bits) for the 3rd coefficient
3 The upper 8 bits U4 of the gain data G (8 bits in total) are taken out for each block, and these upper bits Ll
t to U4 are inserted into the remaining area of the signal Sa.

In this case, the upper bits U1 to U4 are 24 bits per block, or 3 words, so in the field period, 3 words x 26 blocks = 78 words, and these upper bits U1 to U4 are as follows: It can be inserted into the remaining 93 word area.

The above encoding process 7 is performed in the encoder (73), and the encoding process will be explained next.

FIG. 1 shows the data allocation (memory map) in the memory used for the encoding process, and the address has a capacity of 1 word (8 bits). Also, the size of this memory is for one field period, and 13
It is 2 words x 12 words.

Then, at the time of encoding, i, the prediction residual data at-G is 4 times per sample.
In addition to being a bit, it is also 1 bit for the left and right channels.
Since they are paired, the left channel data 5t-c is the top 4
The data 6t-c of the right channel is combined so as to occupy the lower 4 bits, and one data Mi having a length of one word is formed.

In this case, as mentioned above, data (5t-c) is obtained for 801 samples x 2 channels in one field period, and therefore, data Mi is 801 samples in one field period.
Words will be obtained, so these 801 words can be converted into data Mo = Msoo (i = 0 to 8
00).

110 Next, the data Mi is sequentially written into the memory of FIG. 1 at the address locations shown in the figure (.

iii. Also, as mentioned above, during the ■ field period, the data is set to 1 to 3. Since G is obtained in 26 sets of 6 words, each word is converted into data SIIn (m is the word number in the data of 6 words per set, and m = 0 to 5゜n is
Block numbers for 26 sets of data, n=o~25
), this data Smn is sequentially written to address locations as shown in the figure.

iv, 26 sets of data kl~, 3. The upper bits U1 to U4 are extracted from G as shown in FIG. 2, and these are 3 words of data Uin+ U2111U4 for each set
n is reconfigured and written to the memory as shown in the figure.

Note that in this case, actually, the write process of ii = iv Ljl can be performed sequentially for each l block.

v, 6 words of user data (11 words) ID
n=IDs is written as shown in the figure.

For the data written in terms vi, 1ixv, 132 pairs of P parity words Pj (j=O~131
) and Q parity words Qj are generated, and these parity words Pj, Qj are written into each address as shown.

A CRC code Cr (r-0 to 263) is generated for the data of vi, ii = vi terms, and this is written to each address of the memory as shown in the figure. However, in this case, Ct (C code Cr is two words, the even numbered code and the next odd numbered code, which is the original 1
It forms one CRC code.

vw+, when the above writing is completed, the data in the memory in FIG.
from the second column to the 132nd column.
The columns are read in the same way.

However, at the time of reading, as shown in FIG. 3, for each column of the read data, a 3-bit marker MK and an 8-bit column address AD) are placed at the beginning. S
is added. In addition, in the figure, data W is data l Do = I L)5. Mi.

S III U III ” U 4n.

Furthermore, during this readout, time axis compression is performed by increasing the readout speed. Furthermore, since the data write order and the data read order are different, interleaving is performed. This read data string is the signal Sa and is supplied to the modulation circuit (74).

The above-mentioned encoding process is for one field period, and in reality, it is necessary to process the audio signal R for more than one field period.
There are two memories in the figure, and in these two memories, the above-mentioned encoding process is performed every one field period, with a one-field period shift from each other, and the signals Sa are successively taken out.

Note that in the above, data Mi + Smn, Ut
The parity code, CRC code, formant of the signal Sa, etc. are the same as those in the current 8 mm video, except that n to Urn are data based on the predictive coding method.

On the other hand, during decoding, 1. The signal Sa from the demodulation circuit (91) is written to the same address in the memory (FIG. 1) in the same order as read during encoding. However, at this time, marker MK and address ADR3 in signal Sa are not written.

(2) Error data is detected using the P parity word Pj, and among the error flags (error pointers) prepared for each data, the error flag corresponding to the error data is set.

■For data with error flags set,
Error correction is performed by P parity word Pj. When this error correction is correctly performed, the corresponding error flag is reset.

The same processing as in the ■ and ■ terms is performed using the lV and Q parity words Qj.

■If there is an error flag set above, ■
The processing of ~IT/terms is repeated. However, the maximum number of repetitions is, for example, five times.

Therefore, when this section 7 is completed, error data may exist, and the error flag of that error data is set.

Further, the processing up to this point is the same as that for current 8 mm video.

Vl, error flag of data Smn, for example, '! of the first block. When the error flag of 1 is checked for the A1-order prediction coefficient and it is reset, there is no error in 1 for that prediction coefficient, so 1 is read from the memory for that prediction coefficient, and 1 is set for this coefficient. will be served for lunch (51).

■When the error flag is set in V1i[, the error flag of the data corresponding to data S+++n among the data U111 to U411 written in 2*, in the present example, the error flag of the data of the first block. When the error flag of the double-written data Urt for 1 in the first prediction coefficient is checked and reset,
The data U11 is set as 1 in the prediction coefficient (51)
latched to.

At this time, 1 in the prediction coefficient is 16 bits, whereas data U1t is the upper 8 bits of 1 in the coefficient, so
The lower 8 bits are set to "0".

When the error flag is marked in section ■ or ■,
For example, a latch operation of 1 is not performed on the prediction coefficient for the latch (51), and therefore the prediction coefficient kl of the previous block is used as 1 as the prediction coefficient of the current block, that is, the prediction is made by holding the previous value. 1 is interpolated to the coefficient.

The processing in sections IX and Vl to ■ is performed in the same way for all data Smn.

Then, the above processing is performed using two sets of memories as in the case of encoding, with a difference of one field period from each other for each field period, and data t-G, kt~
3. G is decoded, and this decoded data t
At 5t-G, kt~, the data start and start are decoded based on 3+G, and the signal R is taken out.

In this case, when data Snn has an error that cannot be corrected, by the processing in item (2), only the prediction coefficients 1 to 2 or the upper bits of gain data G are changed to data U 1
The lower bits are corrected by U 411, and the lower bits are set to 0'', for example, but the lower bits are less important than the upper bits in one data, so the complete coefficient is set to 1.
Compared to the case where the data 2L is decoded using .about.3 and data G, the error of data x[ increases only slightly, and no audible problem occurs.

Effects of the Invention Thus, according to the present invention, when digital audio data is recorded and reproduced by ADPCM, the prediction residual for each sample and the prediction parameter for each block are recorded, and the prediction parameters with high importance are recorded. Since the most important bits of the prediction parameter are recorded twice, even if an uncorrectable error occurs in the original prediction parameter, the audio signal can be reproduced without any audible problem. can.

Also, instead of recording all of the highly important prediction parameters twice, N
Since only the high-order bits with high quality are recorded in duplicate, the error correction ability for highly important prediction parameters is improved even if the recording capacity is limited.

[Brief explanation of the drawing]

FIG. 1 is a memory map diagram of an example of the present invention, and FIGS. 2 to 8 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 constitute one block, and for each block, a prediction parameter is determined from the samples included in that block, and based on this prediction parameter. Then, a prediction residual is calculated for each sample of the block for which this prediction parameter was calculated, and this prediction residual is sent out at the rate of the sample, and the prediction parameter is sent out at the rate of the block. , A digital audio data encoding method in which the upper bits of parameters with high importance among the prediction parameters are transmitted twice in the proportion of the blocks.