JPH09130258A - Audio recording and reproducing device and audio signal encoding device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce an arithmetic operation amount with simple constitution in obtaining an editing point when editing audio signals. SOLUTION: This device performs an encoding processing to input signals from an input terminal 1, based on a prescribed code and records them on a tape 13. In this case, a decoding part 17 for performing a decoding processing to signals read from the tape 13 based on the prescribed code, a switch 4 for switching the input signals and reproducing signals from the decoding part 17 and an encoding part 11 for encoding the audio signals sent through the switch 4 are provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio recording / reproducing apparatus for recording an audio signal on a recording medium and reproducing the recording medium.

[0002]

2. Description of the Related Art In recent years, in order to reduce the amount of data for digital audio signals, for example, a moving picture coding coordinating organization (moving picture image codin) has been developed.
G experts group (MPEG) standardized by M
Sub-band coding and decoding methods are used in coding algorithms such as PEG-1. This subband encoding / decoding method is a method of reducing the data of a redundant portion included in a stereo audio signal by utilizing the property of stereo audio signal data being unevenly distributed in the frequency axis direction. Is.

In the algorithm of the sub-band coding and decoding system, first, the entire band of the 16-bit linearly quantized input signal is converted to 32 by the sub-band analysis filter.
The band is divided and the subband signal is extracted. The sub-band signal has, for example, 12 samples of data, and the ratio of 1 to 32, that is, the clock rate is reduced to 1/32 of the original.

Next, the data of 12 samples of the sub-band signal of each band is divided into a waveform and a scaling factor and normalized so that the maximum amplitude becomes 1.0, and the scaling factor at this time is used as a scale factor. Taken out. On the other hand, the input signal is fast-Fourier-transformed and the result is used to calculate the masking value.

Then, based on the masking value and the scale factor, a quantization bit for performing a quantization process is assigned to each subband signal. Then, each subband signal is quantized based on this quantized bit.

A bit stream is formed in a predetermined format from the quantized signal, the quantized bit, and the scale factor, and the bit stream is recorded on a storage medium.

In the above-mentioned quantization, that is, in the decoding process of the encoded signal, first, a bit stream is taken out from the storage medium, and the quantized signal, the quantized bit, and the scale are extracted from the bit stream. And the factor is obtained. Subsequently, this quantized, i.e. encoded signal is based on the quantized bits,
It is inversely quantized into a 32-band subband signal. 31 zero values are inserted into these subband signals, an unnecessary spectrum generated by the insertion of these zero values is removed by a bandpass filter, and these signals are filter-synthesized.

In this way, it is possible to reproduce the original audio signal by decoding the encoded audio signal.

FIG. 10 shows an example of the above-mentioned audio recording / reproducing apparatus.

First, an audio signal is input from the input terminal 51. This audio signal is sent to the encoding unit 52, is encoded by a predetermined code, and is recorded by the recording amplifier 53.
At a recordable level and is recorded on the tape 63 by the recording head 64.

Subsequently, the audio signal recorded on the tape 63 is read by the reproducing head 65 and reaches a level at which reproduction can be performed by the reproducing amplifier 56.
7 In the decoding unit 57, the decoding process is performed based on the code when the decoding process is performed, and the decoding process is sent to the first output terminal 59.

The audio signal output from the decoding unit 57 is a high quality signal that is comparable to the audio signal before encoding.

[0013]

By the way, in order to edit the tape 63 on which the audio signal coded by the MPEG coding algorithm is recorded, an editing point which usually consists of, for example, a time code is adopted. . In order to adopt this editing point, high-quality voice obtained by decoding the decoded signal with the MPEG decoding algorithm has been used.

Therefore, even if only the editing points are taken, a high quality sound is used more than necessary, resulting in a high price and a long time for occupying the audio recording / reproducing apparatus, so that other work can be performed during that time. There are some problems that I can't do.

Further, if only the editing points are taken, an audio signal (hereinafter referred to as a simple reproduction signal) obtained by reproducing only a part of the low-frequency side signal, even if the sound is not of high quality, is used. ) Is known to be sufficient.

Therefore, as shown in FIG. 10, a code string conversion circuit 58 to which the output of the reproduction amplifier 56 is sent is provided, and a part of the encoded signal sent from the reproduction amplifier 56 is taken out by this code string conversion circuit 58. It is proposed to perform an operation and send only the low frequency side to the second output terminal 60, for example.

However, even if only the code string conversion circuit 58 is used, the newly added structure may be complicated, and it is desirable to use a simpler structure.

Therefore, the present invention has been made in view of the above-mentioned circumstances, and audio recording that can reduce the amount of calculation with a simple structure when obtaining an editing point at the time of editing an audio signal. An object is to provide a reproducing device.

[0019]

An audio recording / reproducing apparatus according to the present invention is an audio recording / reproducing apparatus which encodes an input audio signal based on a predetermined code and records it on a recording medium. A decoding unit that performs a decoding process on the generated signal based on the predetermined code, a switching unit that switches between the input audio signal and the reproduced audio signal from the decoding unit, and audio that is sent through the switching unit. The above-mentioned problem is solved by including an encoding unit that encodes a signal.

Further, it is possible to provide a thinning means for thinning the output from the decoding section.

Further, according to the above audio recording / reproducing apparatus, the audio signal read from the recording medium is subjected to the decoding processing by the decoding section based on the predetermined code and sent to the switching means. The switching unit switches between the input audio signal and the reproduced audio signal sent from the decoding unit as the audio signal input to the encoding unit. The encoding unit encodes the audio signal sent from the switching unit based on a predetermined code. It should be noted that the encoding unit performs normal recording encoding when an input audio signal is input, and outputs an audio signal having a smaller data amount than that obtained by recording encoding when a reproduction audio signal is input. The resulting encoding process is performed.

Further, by adding the thinning means, when the audio signal once decoded can be coded, it can be input to the coding section in a state in which the data amount is reduced.

Further, the audio signal coding device according to the present invention is an audio signal coding device which performs coding processing with a predetermined code for each signal of each band obtained by dividing the band of an audio signal into M parts. , An audio signal coding apparatus, wherein coding processing is performed using only a part of a band of an M-divided audio signal.

According to the above audio signal encoding device,
Since the encoding process is performed using a part of the input audio signal, the amount of calculation and the processing time required for the encoding process can be reduced.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Specific examples to which the audio signal encoding device and audio recording / reproducing device according to the present invention are applied will be described below with reference to the drawings.

In the first concrete example of the audio recording / reproducing apparatus, as shown in FIG. 1, an audio signal input from the input terminal 1 (hereinafter referred to as an input signal) is encoded based on a predetermined code. In an audio recording / reproducing apparatus for recording on a tape 13, a decoding unit 17 for decoding a signal read from the tape 13 based on the predetermined code, the input signal, and a reproduced audio signal from the decoding unit 17. A switch 4 for switching (hereinafter referred to as a reproduction signal) and a coding unit 11 for coding an audio signal sent via the switch 4 are provided.

The audio signal encoding device is applied to the encoding unit 11. That is, the encoding unit 11
Is a part of the M-divided audio signal band in an audio signal coding device that performs coding processing with a predetermined code for each signal in each band obtained by dividing the audio signal band into M parts. The encoding process is performed only when the selected terminal a is selected by the switch 4 using only the band.

According to FIG. 1, the input signal from the input terminal 1 is, in the encoding section 11, as will be described later, an MPEG (moving pictu) which is an international standard for media integrated video compression.
It is encoded by an encoding algorithm by a re-expert group (hereinafter referred to as normal encoding), further processed by the recording amplifier 12, and recorded on the tape 13 by the recording head 14.

The audio signal recorded on the tape 13 is read by the reproducing head 15 and is reproduced by the reproducing amplifier 1.
6, and the decoding unit 17 performs the decoding process with the algorithm described later. The decoded audio signal is output not only to the first output terminal 2 but also to the selected terminal a of the switch 4.

Here, when the reproduction signal is input by the switching operation of the switch 4, the encoding unit 11 performs a simple encoding process, which will be described later, on the reproduction signal, and the simple encoded audio signal. The signal (hereinafter, simply coded signal) is sent to the second output terminal 3. When an input signal is input, the normal encoding process is performed on the input signal as described above.

The simplified coded signal has a data amount smaller than that of the input signal, as described later.

In FIG. 1, an encoding unit 11 is a unit for encoding the input signal or the reproduction signal by the normal encoding process algorithm or the simple encoding process algorithm, as will be described later. . here,
The encoded audio signal is sent to the recording amplifier 12 and also to the second output terminal 3.

The recording amplifier 12 amplifies the audio signal that has been subjected to the normal encoding process from the encoding unit 11 to raise it to a recordable level, and then outputs it to the recording head 14.
The output audio signal is recorded on the tape 13 by the recording head 14. Further, when the audio signal subjected to the simple coding process is input from the coding unit 11, the operation is stopped.

When the audio signal recorded on the tape 13 is reproduced, after the tape 13 is mounted, the reproducing head 15 reads out the recorded content and the reproducing amplifier 16
To the decoding unit 17 after amplifying the output encoded signal to a level at which reproduction processing can be performed.

As will be described later, the decoding section 17 takes out only the data of a part of the band of the coded signal which has been band-divided and coded and decodes it, and outputs the first output terminal 2 and the switch. 4 to the selected terminal a.

The switch 4 switches the terminals a and b to be selected and selects a signal to be input to the encoding unit 11. here,
The reproduced signal from the decoding unit 17 is sent to the selected terminal a, and the input signal from the input terminal 1 is sent to the selected terminal b.

Although a tape is used as the recording medium here, another type of recording medium such as a disc-shaped recording medium may be used.

Here, the encoding unit 11 to the tape 1 is shown in FIG.
3 shows the signal flow from the tape 13 to the decoding unit 17.

In FIG. 2, the analysis filter bank 10
1, scaling unit 102, bit allocation unit 103,
The operations of the quantizing unit 104 and the formatting unit 105 are
The contents are processed by the encoding unit 11, and the operations in the bitstream expansion unit 107, the dequantization unit 108, and the synthesis filter bank 109 are the contents processed by the decoding unit 17.

According to FIG. 2, the audio signal input terminal 1
The input signal at 00 is band-divided and output at the analysis filter bank 101 of the encoding unit 11 in the normal encoding process. On the other hand, in the simple encoding process, the band is divided and the data amount is reduced and then output. Also,
After the scaling 102 and the bit allocation unit 103 perform a predetermined process on the output signal from the analysis filter bank, a quantization unit 104 performs a quantization process or an encoding process, and a format unit 105 records the signal. It is recorded on the tape 13 as a medium.

On the other hand, the encoded signal recorded on the tape 13 is read by the reproducing head 15 and then output.
After the bit stream expansion unit 107 and the inverse quantization unit 108 of the decoding unit 17 perform predetermined processing, band synthesis is performed by the synthesis filter bank 109. At this time, as will be described later, a reproduced signal having the same data amount as the input signal or a reproduced signal having a smaller data amount than the input signal can be obtained.

In FIG. 2, an audio signal input terminal 1
The output from 00 is input to the encoding unit 11 and is also sent to the analysis filter bank 101.

The analysis filter bank 101 has, for example, a low-pass filter, ie, a so-called low-pass filter, as a band division filter, and divides an input audio signal into M pieces, for example, every predetermined number of samples s. ,
The signal divided for each s sample of each band is sent to the scaling unit 2. Below, this analysis filter bank 101
The following shows a specific example.

First, the signal S [k] obtained after the band division processing by the simple coding processing is defined by the following equation (1).

[0045]

(Equation 1)

Here, the number of taps of the filter used for band division is 256.

Therefore, the first specific example of the analysis filter bank 101 at the time of the simple coding process is, as will be described later, a 1 on the lower low frequency side of the band obtained by dividing the input signal into 32, for example.
Only the 6 bands are extracted and the signals S [k] for 16 bands are extracted.
(0 ≦ k <16) is output.

On the other hand, in the second specific example of the analysis filter bank at the time of the simple encoding process, as will be described later, the input signal is divided into, for example, 16 parts, and all the subband signals obtained by the 16 parts are extracted. Then, the signals S [k] for these 16 bands
(0 ≦ k <16) is output.

Therefore, prior to the description of the operation of each specific example of the analysis filter bank 101 during the simple coding process, a conventionally known band division filter during the normal coding process will be described.

FIG. 3 shows a flow chart for explaining the operation of the band division filter bank during the normal encoding process. Further, FIG. 3 shows an example in which the number of taps is 512.

In step S101, for the input signal (X [i]), as shown in equation (2), 3
A so-called shift processing operation is performed to divide the taps every second tap, and the process proceeds to step S102.

X [i] = X [i-32], where 32 ≦ i <512 (2) Further, in step S102, the signals of 32 samples shifted in step S101 are input.

Here, S [k] which is the sub-band signal after the target band division is defined by using X [i] as shown in equation (3). Note that F [i] in the equation (3) is a coefficient of the prototype low-pass filter, and M [i] [k] is this F
It is a rotor of [i] and is defined by equation (4). Further, the rotor M [i] [k] exhibits the property shown in the equation (5). That is, it shows the property that the sign is inverted while the size is kept every time the tap advances by 64.

[0054]

(Equation 2)

In step S103, so-called filter processing is performed by multiplying X [i] by the filter coefficient F [i], as shown in equation (6), and 512 signals Z [i] (0≤i
<512) is obtained. Here, F [i] (0 ≦ i <5
Instead of 12), C [i] (0 ≦
i <512) is used. According to the equation (7), C [i]
Indicates that the sign of F [i] is inverted in units of 64 while keeping the size of F [i] as it is.

Z [i] = C [i] × X [i], where 0 ≦ i <512 (6) C [i + 64j] = (-1) ^j F [i + 64j], where 0 ≦ i <64, 0 ≦ j <8 (7) In step S104, Z obtained in step S103
Since [i] is a value in which j is mixed, in order to eliminate this j, as shown in equation (8), the sum of Z [i] for each tap, a so-called partial sum, is calculated, and Y [i] Is calculated.

[0057]

(Equation 3)

In step S105, the product of Y [i] and the rotor M [i] [k] is used to satisfy 0≤i, as shown in equation (9).
In the range of <64, the sum is taken to obtain S [k].

[0059]

(Equation 4)

In step S106, the obtained 32 signals S [k] (0≤k <32) or subband signals are output.

Based on the above, the operation of the analysis filter bank 101 during the simple encoding process will be described.

First, in the first specific example of the analysis filter bank during the simple encoding process, the input signal is sampled every 32 taps in steps S101 and S102 as described above, and is sampled in step S103. Based on the selected subband signals, 256 Z [i] are calculated in the range of 0 ≦ i <256 as shown in the following expression (10). Further, in step S104, Y [i] is obtained by adding Z [i] obtained in step S103 for each i as shown in the following equation (11). Also,
In step S105, 16 subband signals S [k] (0 ≦ k <16) are calculated as shown in the following equation (12), and this S [k] is output in step S106.

[0063]

(Equation 5)

That is, this analysis filter bank 101
In the first specific example of (1), the amount of calculation and the amount of output data are reduced by performing signal processing for a part of the band of the band-divided subband signal. A second specific example of the analysis filter bank 101 will be described as a specific example of reducing the calculation amount and the output data amount.

Here, in the second specific example of the analysis filter bank 101 at the time of the simple encoding process, the input signal is divided into 16 parts, for example, and all the sub-band signals obtained by the 16 parts are processed. Therefore, the rotor M [i] [k] at this time is defined as shown in the following equation (13),
As shown in the equation (14), it has a property that the sign is inverted every 32 pieces.

M [i] [k] = cos ((i-8) (2k + 1) π / 32) (13) M [i] [k] = − M [i + 32] [k ] (14) That is, the shift process is performed in step S101.
The taps are separated for each, and in step S102, the signals of the remaining 16 samples are input and divided into 16 bands. In the filtering process in step S103, Z [i] is calculated as shown in the following expression (15). here,
C [i] (0 ≦ i <256) in the equation (15) is given by (16)
As shown in the equation, the sign of the coefficient F [i] of the prototype low-pass filter used as the band division filter is inverted in units of 32.

Z [i] = C [i] × X [i], where 0 ≦ i <256 (15) C [i + 32j] = (-1) ^j F [i + 32j], 0 ≦ i <32, 0 ≦ j <8 (16) In step S104, the partial sum Y [for each i for Z [i] obtained in step S103 is expressed as shown in the following expression (17). i] is required. Further, step S
In the band division processing at 105, the partial sum Y [i] is multiplied by the rotor M [i] [k] and added for i to calculate S [k] as shown in the following equation (18). And step S10
It is output at 6.

[0068]

(Equation 6)

Returning to FIG. 2, the scaling unit 102 obtains the maximum amplitude, that is, a so-called scale factor for each signal for each band, that is, for each s sample, and sends the scale factor and the signal for each band to the bit allocation unit 103. In addition, the bit allocation unit 103 allocates a quantized bit for each band based on the scale factor and quantizes the quantized bit and the signal for each band.
Send to Further, the quantization unit 104 quantizes the signal in each band based on the quantized bit, and sends the quantized signal to the bitstream format unit 105.

Also, the bit stream format section 1
05 forms a bit stream from the quantized signal and the quantized bit, and records the bit stream on the tape 13.

The decoding process will be described below.

The data recorded on the tape 13 is read by the reproducing head 15 and sent to the decoding section 17.

Therefore, as a concrete example of the decoding unit 17, M
An example of a decoding process using a decoding algorithm in PEG (hereinafter referred to as a normal decoding process), and two examples of a simple decoding process that can obtain a reproduction signal having a smaller data amount than the reproduction signal obtained by the normal decoding process. Will be described.

The first specific example of the decoding unit 17 at the time of simple decoding is the inverse quantization of a coded signal of a part of the coded signal recorded on the tape 13, for example, the low frequency side 16 band. And a filtering process is performed to obtain a reproduced signal.

On the other hand, the second specific example of the decoding unit 17 is composed of a low-pass filter having a width of π / K and a center frequency of an odd multiple of π / 2K as the synthesis filter bank 109 as described later. A band pass filter, a so-called band pass filter is used, and the inverse quantizer 10
If the target number of sub-bands M, for example M = 32, is output when the signal after the decoding processing in 8 is performed, for the data of a part of the band of the encoded signal recorded on the tape 13, (K-
1) For example, K = 16, that is, (16-1) = 15 zero values are inserted, and a K band obtained by performing band synthesis in the synthesis filter bank 109, in this case, a reproduction signal for 16 bands is obtained. It is a thing. Note that K is a value that satisfies the relationship of 1 ≦ K <M.

Therefore, first, in the first concrete example of the decoding unit 17 at the time of the simple decoding process, the bit stream expansion unit 107 extracts, for example, 3 bits from the transmitted bit stream.
The quantized bits for two bands, the scale factor, and the quantized signal, that is, the coded signal are extracted, and each data is sent to the inverse quantization unit 108.

The inverse quantization unit 108 takes out, for example, a signal in the lower 16 bands from the 32 band encoded signal sent from the bit stream expansion unit 107, performs inverse quantization, and sends it to the synthesis filter bank 109.

The synthesis filter bank 109 has, for example, 25
A part including a 6-tap low-pass filter,
The 16-band inverse-quantized signals sent from the inverse-quantization unit 108 are band-combined to output a reproduction signal.

Here, prior to the description of the operation of the synthesis filter bank 109, the operation of the band synthesis filter during the normal decoding process will be described.

First, FIG. 4 shows the operation of the band synthesizing filter during the normal decoding process. Note that, here, it is assumed that 512 taps are used for filtering.

In step S110, the inverse quantizer 1
For example, the signal S [k] dequantized from 08 is input. Further, in step S111, so-called shift processing is performed to divide the filter taps into 64 units.

Here, the filter coefficient of the band-pass filter of the k-th band counting from the low band necessary for filtering the dequantized signal is d [i] [k] (0≤i <51
If 2,0 ≦ k ″ <32), the signal after the filter processing has X satisfying the following expression (19).

[0083]

(Equation 7)

Here, the coefficient d [i] [k] of the k-th band-pass filter is the coefficient of the prototype low-pass filter constituting this band-pass filter H [i] (0≤i <
512), and H [i] is used as a rotor M [i] [k] for forming a bandpass filter, d [i] [k] is expressed by the following equation (20). In equation (20), M
[i] [k] is defined by the equation (21) and has symmetry as shown in the equation (22).

D [i] [k] = M [i] [k] × H [i] (20) M [i] [k] = cos ((16 + i) (2k + 1) π / 64) (21) M [i] [k] =-M [i + 64] [k] (22) Further, according to the equation (22), the rotor M [i] [ k], i is 6
It can be said that it has the property of becoming one cycle every four steps.

Further, in order to utilize this symmetry, a filter coefficient D [i] (0≤i <512) in which the sign of the coefficient H [i] of the filter is inverted in units of 64 i is set. And Therefore, this D [i] has the property as shown in the following equation (23).

D [m + 64p] = (− 1) ^p H [m + 64p] (0 ≦ m <64, 0 ≦ p <8) (23) Further, in order to simplify the equation, In the rotor M [i] [k], a rotor whose sign is inverted in units of 64 is N [i] [k].
And Also, this N [i] [k] is (2
It has the property shown in the equation (4). Thus, the filter coefficient D [i] (0≤i <512) and the rotor N [i] [k] are obtained.

N [m + 64p] [k] = (-1) ^p M [m + 64p] [k] (0 ≦ m <64, 0 ≦ p <8) (24) Therefore, (19) The expression () can be modified as shown in the following expression (25).

[0089]

(Equation 8)

Returning to FIG. 2, in step S112, the calculation of "N [i] [k] × S [k]", that is, matrix processing is performed and the number of bands k is erased, as shown in the following equation (26). A signal V [i] (0 ≦ i <512) obtained by the above is obtained.

[0091]

(Equation 9)

Considering that V [i] reaches a cycle every time i advances 64, there are 8 cycles in 512 taps. In addition, the fact that the cycle comes every time i advances 64 means that V [i] is inverted every time i advances 32. Therefore, it is necessary to perform the filtering process separately for the first half and the second half of each cycle. In FIG. 5, this necessary portion is shown by hatching. In addition, i is 0 to 6 in the shaded area.
FIG. 6 shows the gears that have been shifted so as to enter the range 4.

In order to extract the first half and the second half separately, as shown in FIG. 6, it is necessary to extract 16 cycles of data, which is twice as many as 8 cycles.
This corresponds to data for 24 taps. However, this 10
Of the data for 24 taps, the half required for processing is 5
Since there are 12 taps, in FIG.
Processing for obtaining the signal U [i] (0 ≦ i <512) as shown in FIG.

In step S113, if the ordinal number of the cycle in FIG. 7 below is even-numbered, it is processed according to the following expression (27), and if it is odd-numbered, it is processed according to the following expression (28), and U [ i] is taken out.

U [i] = U [r + 64q] = V [r + 128q], where 0 ≦ r <32, 0 ≦ q <8 (27) U [i] = U [r + 32 + 64q] = V [r + 96 + 128q], where 0 ≦ r <32, 0 ≦ q <8 (28) In step S114, each i is represented by the formula (29).
Is multiplied by U [i] and D [i], that is, filtered to obtain a signal W [i].

W [i] = U [i] × D [i] (29) Furthermore, in step S115, W [i] is used to calculate (3
As shown in equation (0), i is added from 0 to 511 to generate X, that is, band synthesis processing is performed, and the reproduction signal X obtained in step S116 is output.

[0097]

(Equation 10)

In consideration of the above, the synthesis filter bank 109 in the simple decoding process of the first specific example of the decoding unit 17 will be described.

In step S110, the inverse quantizer 1
The signal sent from 08 is an n-band signal, for example, n = 16, that is, a 16-band signal. Therefore, step S in FIG.
At 110, the 16-band dequantized signal is input. Therefore, the calculation shown below is as shown in Expression (31).

[0100]

[Equation 11]

In step S111, as described above, 64
The taps are separated for each, and in the same matrix processing as that described above performed in step S112, V [i] is obtained by summing 16 bands, that is, 0 ≦ k <15, as shown in the following expression (32). Furthermore, in step S113,
As shown in Expressions (33) and (34), the preprocessing for the filter processing as described above is performed, and step S1
In 14, the filter processing is performed for each i as shown in the equation (35).

[0102]

(Equation 12)

In step S115, band combining processing is performed as shown in the following equation (36), and step S11 is performed.
At 6, the reproduced signal X, which is an audio signal for 32 samples obtained by this band synthesis processing, is output.

[0104]

(Equation 13)

As described above, according to the first specific example of the decoding unit 17 in the simple decoding process, the coded signal coded with the same number of bands as the number of bands required for the normal decoding process. In order to obtain the reproduced signal by performing the decoding process using a part of the band, the decoding process, that is, the normal decoding process is performed on the coded signal that has been coded using the entire band of the normal audio signal. It is possible to reduce the amount of calculation.

Next, a second specific example of the decoding section 17 will be described.

In the decoding unit 17 of the second specific example, the bit stream expansion unit 107 performs the same operation as described in the first specific example of the decoding unit 17 described above.

Further, the inverse quantizing unit 108 has K (1 ≦ K <M from the low frequency side of the signal transmitted from the bit stream expanding unit 7 with respect to the encoded signal transmitted from the reproducing amplifier 16. , M = 32), for example, 16 bands are selected, and (K-1) zero values are inserted into the coded signals of the selected K bands to perform inverse quantization.

The synthesis filter bank 9 is a band-pass filter composed of a low-pass filter having a width of π / K and a center frequency that is an odd multiple of π / 2K as described above, and is sent from the inverse quantizer 8. The dequantized signals for K bands are band-combined to form a reproduction signal for K bands, in this case, a reproduction signal for 16 bands, and the reproduction signal is sent to the audio signal output terminal 110.

As described above, according to the second specific example of the decoding unit 17 in the simple decoding process, the coded signal coded with the same number of bands as the number of bands required for the normal decoding process. Since a decoding signal is obtained by using a part of the band of (3) to obtain a reproduced signal, the amount of calculation can be reduced and the amount of output data can be reduced as compared with the case of performing the normal decoding process.

By changing the number of zero values inserted in the signal to be decoded, the clock rate can be arbitrarily changed, that is, the sampling frequency can be changed.

According to the first concrete example of the audio recording / reproducing apparatus shown in FIG. 1, by using the first concrete example of the decoding section 17 during the simple decoding process, the number of taps of the normal filter is reduced. Compared with this, it is possible to use a filter with a low-order tap number, so that the amount of calculation and the processing time are reduced. Further, at the time of the simple encoding process using this audio signal, the first specific example of the analysis filter bank 101 is used to divide the band of the input audio signal to extract a part of the band, for example, 16 bands, By performing the encoding process, the amount of data is reduced and the number of calculations and the processing time are reduced.

Alternatively, in the simple decoding process, the second concrete example of the decoding unit 17 is used to obtain an audio signal of 16 bands from an encoded signal of 16 bands, and
Since it is possible to use a filter having a tap number lower than that of a normal filter, the amount of calculation and the processing time are reduced. Further, at the time of the simple encoding process using this reproduced signal, the encoding unit 11 uses the second specific example of the analysis filter bank 101 to perform band division using the signals for the 16 bands, and performs the encoding process. By doing so, the number of calculations and the processing time are reduced as compared with the normal processing.

The present invention is not applied only as an audio recording / reproducing apparatus for recording and reproducing an audio signal, but is used as a recording apparatus only for recording on a recording medium, or is recorded by an ordinary recording method. It may be used as a reproducing device for reproducing a recording medium.

In the second specific example of the audio recording / reproducing apparatus, as shown in FIG. 8, an audio signal input from the input terminal 1 (hereinafter referred to as an input signal) is encoded based on a predetermined code. In the audio recording / reproducing apparatus for recording on the tape 13 as a tape, a decoding unit 17 for decoding a signal read from the tape 13 based on the predetermined code, the input signal, and a reproduction audio from the decoding unit 17. A switch 4 for switching a signal (hereinafter referred to as a reproduction signal) and a coding unit 11 for coding an audio signal sent via the switch 4 are provided.

It is also possible to include a thinning circuit 19 for thinning the output from the decoding unit 17.

According to the second specific example of the audio recording / reproducing apparatus, the decoding unit 1 is used, for example, during the simple decoding process.
The synthesis filter bank 109 of the first example of No. 7 is used. That is, the coded signal divided into a smaller number of bands than the normal process and coded is subjected to the decoding process using a band synthesis filter having a low order to obtain a reproduced signal. Further, the reproduction signal is sent to the thinning circuit 19, and the thinning circuit 19 thins out the data every other sample, for example. In this way, the amount of data is reduced. The output from the thinning circuit 19 is output to the selected terminal a of the switch 4 and is input to the encoding unit 11 according to the operation of the switch 4.

Also, during the simple coding process, the coding unit 11 performs the coding process using the second example of the analysis filter bank 101.

Further, in FIG. 8, the same numbers as the numbers given to the configuration shown in FIG. 1 perform the same operations as the operations of the above-mentioned respective parts, and therefore the description thereof will be omitted here.

The decimating circuit 19 decimates the reproduced signal sent from the decoding section 17 at every other sample, for example, to obtain an audio signal having the original half number of samples. That is, the number of samples of the audio signal output from the decoding unit 17 is reduced.

As described above, in order to reproduce the re-encoded coded signal, the decoding section 17 may be used as it is. However, for example, the coded signal is output to the second output terminal 3. A playback device dedicated to the output and the simple decoding process may be used.

As shown in FIG. 9, the reproducing apparatus dedicated to the simple decoding process processes the re-encoded coded signal sent from the input terminal 71 in the recording amplifier 72 and then records it in the recording medium. Data is temporarily recorded on a disk 79 using the recording head 14. In addition, the playback head 75 is used for the disc 7
The signal read from 9 is sent to the decoding unit 77 through the reproduction amplifier 76, the decoding unit 77 performs the above-described simple decoding process, and the obtained signal is sent to the output terminal 78.

The output of the playback device dedicated to the simple decoding process is used, for example, to select an edit point for audio editing.

Note that the present invention is not applied only as an audio recording / reproducing apparatus for recording and reproducing an audio signal, but is used as a recording apparatus only for recording on a recording medium, or is recorded by an ordinary recording method. It may be used as a reproducing device for reproducing a recording medium.

[0125] Further, as an example of the use of the reproduction signal subjected to the simple decoding process, an example of using it for selecting an editing point at the time of audio editing was given, but the present invention is not limited to this, and a high quality audio signal is required. The effects of the present invention can be obtained in any applications that do not

Further, as the data recorded on the tape 13, the data divided into 32 bands and encoded is used, but the data obtained by dividing into another band number may be used. . In the decoding unit 17, an example in which the low-frequency side 16 bands are extracted from the encoded signal extracted from the tape 13 has been described, but the present invention can be achieved even if other parts are extracted in the range in which only a part of the bands is handled. The effect of can be obtained.

[0127]

As described above, according to the audio recording / reproducing apparatus of the present invention, when the signal recorded on the recording medium is decoded and re-encoded, the data amount is set to the above recording medium. In addition to reducing the number of signals recorded in, and using a band division filter or band synthesis filter of a lower order than that used during normal encoding and decoding processing, the amount of calculation by these filter processing is reduced. Therefore, it is possible to shorten the processing time.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first specific example of an audio recording / reproducing apparatus according to the present invention.

FIG. 2 is a block diagram illustrating signal processing of an encoding unit and a decoding unit of a first specific example of the audio recording / reproducing device.

FIG. 3 is a diagram showing a specific example of the operation of a band division filter used in the encoding unit.

FIG. 4 is a diagram showing a specific example of an operation of a band synthesizing filter used in the decoding unit.

FIG. 5 is a diagram visually showing an operation in the middle of the band synthesis filter at the time of normal decoding processing.

FIG. 6 is a diagram visually showing an operation in the middle of the band synthesis filter during the normal decoding process.

FIG. 7 is a diagram visually showing an operation in the middle of the band synthesis filter during the normal decoding process.

FIG. 8 is a block diagram showing a configuration of a second specific example of the audio recording / reproducing apparatus.

FIG. 9 is a block diagram showing a configuration of a disc recording / reproducing apparatus for recording on a recording medium by using a signal encoded by the conventional audio recording / reproducing apparatus.

FIG. 10 is a block diagram showing a configuration of a conventional audio recording / reproducing apparatus.

[Explanation of symbols]

 11 Encoding Unit 13 Tape 14 Recording Head 15 Playback Head 17 Decoding Unit 19 Decimation Circuit 101 Analysis Filter Bank 108 Inverse Quantization Unit 109 Synthesis Filter Bank

Claims (9)

[Claims] 1. An audio recording / reproducing apparatus which encodes an input audio signal based on a predetermined code and records it on a recording medium, wherein a signal read from the recording medium is decoded based on the predetermined code. A decoding unit, a switching unit that switches between the input audio signal and a reproduced audio signal from the decoding unit, and an encoding unit that encodes an audio signal sent through the switching unit. Audio recording and reproducing device. 2. The audio recording / reproducing apparatus according to claim 1, further comprising thinning means for thinning the output from the decoding section. 3. The encoding unit performs an encoding process using only an audio signal in a part of a band of the audio signal divided into M, according to the switching operation of the switching unit, and 2. The audio recording / reproducing apparatus according to claim 1, wherein an operation of performing an encoding process using the entire band of the audio signal divided into M is switched. 4. The encoding unit performs band division using a band division filter of a lower order than the order required for band division using the entire band of the audio signal, according to the switching operation of the switching unit. 2. The audio recording / reproducing apparatus according to claim 1, wherein the operation for performing the band division and the operation for dividing the band using a normal order filter are switched. 5. The operation of the decoding unit performing a decoding process using only a part of the band of an encoded signal which has been M-divided and encoded according to the switching operation of the switching means, 2. The audio recording / reproducing apparatus according to claim 1, wherein an operation of decoding using the entire band of the coded signal that has been M-divided and coded is switched. 6. The decoding unit, in accordance with the switching operation of the switching means, performs a decoding process using a band synthesis filter of a lower order than the order required for band synthesis of all bands of the encoded signal. 2. The audio recording / reproducing apparatus according to claim 1, wherein an operation for performing the decoding process is switched to an operation for performing a decoding process using a band synthesizing filter required for band synthesizing all bands of the encoded signal. 7. An audio signal coding apparatus for performing coding processing with a predetermined code for each signal of each band obtained by dividing the band of an audio signal into M, of the band of the audio signal divided into M An audio signal encoding device, which performs an encoding process using only a part of the band. 8. The audio signal encoding device according to claim 7, wherein the band selected by the band division processing of the audio signal is a band on the lower side of a predetermined band. 9. The band splitting process for the audio signal, wherein a band splitting filter having a lower order than the order required for the band splitting process for the entire band of the audio signal is used. Audio signal encoding device.