JPH10294668A - Method, device for decoding audio encoded data and record medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decode audio encoded data at optimum high speed by lightening a load for decoding corresponding to conditions such as the reproducing ability of the reproducing device and the throughput of the computer. SOLUTION: An ability analysis control means 5 judges the reproducing ability of the reproducing device, the throughput of the computer and the setting of a user, generates a band control signal, a scale factor comparison control signal and a status setting control signal and applies these signals to a bit allocation information extracting means 6 and a scale factor information extracting means 7. Based on the band control signal and scale factor comparison control signal, the scale factor information extracting means 7 reloads and rearranges scale factor information. An inverse quantizing means 3 performs the inverse quantization of the quantized band dividing signal. A band synthesizing processing audio data generating means 4 reproduces audio data corresponding to the signal throughput.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to MPEG (Mov).
ing Picture Experts Grou
p) Regarding decoding of audio encoded data which has been subjected to band division encoding such as an audio standard, particularly describes an audio encoded data decoding method, an audio encoded data decoding apparatus, and a program of an audio encoded data decoding method. And a storage medium.

[0002]

2. Description of the Related Art As a conventional technique for decoding audio encoded data, a decoding method according to the MPEG audio standard will be described as a first conventional example with reference to the drawings.

FIG. 17 is a flowchart showing a signal processing procedure in a decoding method according to the MPEG audio standard. In FIG. 17, in step S01 (hereinafter, notation of steps is omitted), audio encoded data of the MPEG audio standard is input, and audio encoded data input processing is performed. The header information extraction / analysis processing of S02 is for extracting and analyzing header information from audio encoded data of the MPEG audio standard input by the audio encoded data input processing of S01.

The bit allocation information extraction processing of S03 is for extracting bit allocation information from header information of audio encoded data of the MPEG audio standard. The scale factor information extraction processing of S04 is for extracting scale factor information from header information of encoded audio data of the MPEG audio standard. The quantization band division signal extraction processing in S05 is as follows.
It extracts a quantized band division signal from audio encoded data of the MPEG audio standard.

[0005] The inverse quantization process in S06 is for performing inverse quantization of the quantized band division signal to generate a band division signal. Details of the inverse quantization process in S06 are shown in the flowchart of FIG.

The band synthesis processing / audio data output processing in S07 is for band-synthesizing the band-divided signal, converting the band-divided signal into audio data, and outputting the audio data. The details of the band synthesis processing and audio data output processing in S07 are shown in the flowchart of FIG.

The details of the inverse quantization processing of the decoding method in S06 will be described with reference to the flowchart of FIG. S
08 band number upper limit setting processing is based on the header information extracted / analyzed in the header information extraction / analysis processing of S02.
The upper limit of the number of bands is set according to the bit rate, sampling rate, and layer.

[0008] The inverse quantization information reading process of S09 is for reading information necessary for the inverse quantization that is defined in advance. The scale factor reading process in S10 reads the contents of the scale factor extracted in the scale factor information extraction process in S04. The quantization band division signal readout processing in S11 is for reading out the quantization band division signal extracted in the quantization band division signal extraction processing in S05.

The inverse quantization process of S12 uses the inverse quantization information read by the inverse quantization information read process of S09, and uses the quantization band read by the quantization band division signal read process of S11. Inverse quantization of the divided signal is performed, and S13
Is given to the scale factor multiplication process. S1
The scale factor multiplication process of No. 3 multiplies the signal output by the inverse quantization process of S12 by a scale factor to calculate a band division signal.

The sample number upper limit determination process in S14 determines whether the sample number is the upper limit. If the currently read data is not the upper limit of the number of samples, the process returns to S11. Note that the upper limit of the number of samples differs depending on the layer of the MPEG audio standard. The bandwidth update processing of S15 is to update the bandwidth currently being processed to the bandwidth to be processed next. The band number upper limit determination process in S16 is for determining whether or not the set band number is the upper limit. If it is not the upper limit of the number of bands, the process returns to S09, and the same process is repeated up to the upper limit.

The details of the band synthesis processing and audio data output processing in S07 will be described with reference to the flowchart of FIG. The band division signal input processing of S17 is
6 receives the band division signal generated by the inverse quantization processing. The shift processing of S18 is to move the vector V used in the band synthesis in the time direction. S
The 18 shift process S18 will be described later in detail with reference to FIG.

The matrix operation processing in S19 is performed by converting the vector V shifted in the shift processing in S18 from V0 to V63.
Up to the following equation (1)

(Equation 1) It is calculated by: FIG. 20 to be described later shows a detailed flowchart of the matrix calculation process in S19.

In the vector arrangement processing in S20, the arrangement of the vector V calculated by the matrix operation processing in S19 is changed using the following equation (2).

(Equation 2) Then, the vector U is generated using the following equation (3).

(Equation 3) The vector arrangement processing in S20 will be described later in detail with reference to FIG.

The window calculation process in S21 is to calculate the vector W by the following equation (4).

(Equation 4) D in the equation (4) is a window function, and is defined by ISO / IEC111.
72-3 (Information technology)
gy -Coding of moving pict
ures and associated audio
for digital storage mediai
a at up to about 1.5 Mbit /
s-Part 3: Audio): This is shown on pages 50 to 52 of 1993 (E).

The audio data calculation process in S22 is performed in S
The vector W output from the window operation process 21 is input, and audio data is calculated by the following equation (5).

(Equation 5)

The audio data output process in S23
It outputs the audio data calculated by the audio data calculation processing of No. 22. The one-frame end determination process in S24 determines whether or not the calculation for one frame of the audio encoded data has been completed. If the calculation has not been completed, the processing from S17 onward is repeated.

FIG. 28 is a layout diagram of the vector V before and after the shift processing. In the figure, A01 indicates the array of the vector V before the shift process is performed by the shift process of S18, and the numerical value following V indicates the array number of the vector V. A02 indicates the array of the vector V after the shift process is performed from the shift process of S18, and the numerical value following the V indicates the array number of the vector V in the same manner.

Here, the shift processing in S18 will be described in detail. The array A01 of the vector V before the shift processing includes 1024 vectors V. In addition, the vectors V are arranged in time, and a vector having a larger array number is a past vector. Now, assuming that the shift processing has been performed, 1024 vectors V are 6 in the time direction.
Move four. Like the array A02 after the shift processing, the vector V moves to the right and the past vector V, that is, V960 to V10 of the array A01 of the vector V before the shift processing.
23 is deleted. Further, V0 to V63 of the array A02 of the vector V after the shift processing are set by the matrix operation processing of S19. Also, the above-described equation (2) is used as an arithmetic equation for shifting the vector V.

FIG. 2 shows the details of the matrix operation processing in S19.
0 will be described. The column number setting process in S25 is for setting the column number. Ni of S26
The calculation process of k * Sk is to multiply Nik by the band division signal Sk as shown in equation (1). The Vi calculation process in S27 is performed by accumulating Nik * Sk to obtain the vector Vi.
Is calculated.

The Vi storage process of S28 is to arrange and store the vectors V determined to be the upper limit in the band number upper limit determination process of S16. The column updating process in S29 is to update the column currently being processed to the column to be processed next. The 64 column end determination process of S30 is a process of determining whether the number of columns updated in the column update process of S29 is 64 or more and less than 64. If it is less than 64, the process returns to S26 and thereafter, and the calculation process of Nik * Sk is executed. S
If the number of columns is 64 or more in the 30-column 64-column end determination process, the matrix calculation is terminated.

Next, the details of the vector arrangement processing in S20 will be described with reference to FIG. A03 in FIG. 29 is a layout diagram of the vector V before the vector layout, and A04 is a layout diagram of the vector U after the vector layout. The numerical value after V indicates the array number of the vector V, and the numerical value after U indicates the array number of the vector U.

Array A03 of vector V before vector arrangement
Is composed of 1024 vectors V, and the vectors V are arranged in time. Now, assuming that the vector arrangement is performed, the vector V is converted to the vector U using the equation (3).
It is converted and arranged. After the vector arrangement, 512 vectors U are generated.

Next, a general decoding operation of the encoded audio data will be described. First, S01 shown in FIG.
The audio encoded data is input by the audio encoded data input processing. From the encoded audio data, header information is extracted / analyzed by the header information extraction / analysis processing in S02. After that, the bit allocation information and the scale factor information are extracted by the bit allocation information extraction processing of S03 and the scale factor information extraction processing of S04, respectively. Through the header information extraction / analysis processing in S02, the bit allocation information extraction processing in S03, and the scale factor information extraction processing in S04, information necessary for decoding added by the encoder can be obtained.

Thereafter, a quantization band division signal is extracted from the encoded audio data by the quantization band division signal extraction processing of S05. The quantized band division signal is inversely quantized by the inverse quantization processing in S06 to generate a band division signal. In the inverse quantization process in S06, the process shown in the flowchart of FIG. 18 is performed.

First, in the band number upper limit setting process in S08,
The bit rate, sampling rate, and layer information are read from the header information extracted and analyzed in the header information extraction / analysis processing in S02, and the upper limit of the number of bands is set from the information. Subsequently, in the inverse quantization information reading process in S09, information necessary for the inverse quantization defined in advance is read. Then, by the scale factor reading process of S10, the scale factor extracted in the scale factor information extracting process of S04 is read. Next, by the quantization band division signal reading process of S11, the quantization band division signal extracted in the quantization band division signal extraction process of S05 is read.

Thereafter, using the inverse quantization information read by the inverse quantization information reading process of S09, the inverse quantization process of S12 performs inverse quantization of the quantized band division signal. S
In the scale factor multiplication process of No. 13, the dequantized signal is multiplied by the scale factor read in the scale factor information extraction process of S04 to generate a band division signal.

From the quantized sample readout processing of S11, S1
The process up to the scale factor multiplication process of 3 is repeated for the number of samples, then the bandwidth is updated by the bandwidth update process of S15, and the inverse quantization of S09 is performed until the bandwidth upper limit determination process of S16 determines that the number of bands is the upper limit. The processing from the information reading processing to the band updating processing in S15 is repeated. By such processing, a band division signal is obtained.

Subsequently, the band division signal generated by the inverse quantization processing of S06 shown in FIG. 17 is synthesized with audio data by the band synthesis processing / audio data output processing of S07 and output. In the band synthesis processing / audio data output processing in S07, the processing shown in FIG. 19 is performed.

First, in the band number upper limit setting process in S08,
The bit rate, sampling rate, and layer information are read from the header information extracted and analyzed in the header information extraction / analysis processing of S02, and the upper limit of the number of bands is set. Subsequently, by the band division signal input processing of S17,
The band division signal generated by the inverse quantization processing in S06 is input. This band-divided signal is shifted in S18, S19
Matrix operation processing of S20, vector arrangement processing of S20, S
By performing the window calculation process 21 and the audio data calculation process in S22, it is possible to synthesize the audio data. The audio data output process in S23 outputs the audio data synthesized in this manner.

Since the band synthesis processing and the audio data output processing in S07 are performed in units of one frame,
The processing from the band division signal input processing in S17 to the audio data output processing in S23 is repeated for one frame. As described above, the encoded audio data of the MPEG audio standard is decoded into audio data.

Next, as a conventional technique of the audio encoded data decoding method, there is a "compressed audio information decompression method" disclosed in Japanese Patent Application Laid-Open No. Hei 8-251032. With this as a second conventional example, the decoding method will be described with reference to the drawings.

In the audio coded data decoding method, in the decoding of audio according to the MPEG audio standard, extraction / arrangement of a quantized band division signal of an effective band and extraction / arrangement of scale factor information from bit allocation information. And extraction / arrangement of matrix operation coefficients, and selection / arrangement of inverse quantization information, and perform operations using the arranged information (strictly, rearrangement), thereby speeding up the decoding process. It is.

FIG. 21 is a flowchart showing signal processing of the audio encoded data decoding method in the second conventional example. The same processes as those in the first conventional example are denoted by the same reference numerals, and description thereof is omitted. In FIG. 21, the bit allocation information extraction / arrangement processing in S31 is performed by the MP input from the audio coded data input processing in S01.
It extracts and arranges bit allocation information of audio encoded data of the EG audio standard. S
FIG. 22 shows a detailed flowchart of the bit allocation information extraction / placement processing of No. 31.

S32: Scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection /
The arrangement processing includes extraction / selection / arrangement of scale factor information of audio encoded data of the MPEG audio standard;
And the selection / arrangement of matrix operation coefficients and the selection / arrangement of inverse quantization information. FIG. 23 shows a detailed flowchart of the scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection / arrangement processing of S32.

The quantization band division signal extraction / arrangement processing of S33 is for extracting / arrangement of the quantization band division signal of the audio encoded data of the MPEG audio standard. FIG. 24 shows a detailed flowchart of the quantization band division signal extraction / arrangement processing in S33.

In the inverse quantization process in S34, the quantized band division signal is inversely quantized to generate a band division signal. FIG. 25 shows a detailed flowchart of the inverse quantization process in S34.

The band synthesizing process / audio data output process in S35 is for band-synthesizing the band-divided signal, converting the band-divided signal into audio data, and outputting the data. FIG. 26, which will be described later, shows a detailed flowchart of the band synthesis processing / audio data output processing in S35.

Extraction of bit allocation information in S31 /
Details of the arrangement processing will be described with reference to the flowchart in FIG. The bit allocation extraction processing in S36 is for extracting the bit allocation of audio encoded data of the MPEG audio standard. S3
In the bit allocation arrangement processing of No. 7, the bit allocation extracted in the bit allocation extraction processing of S36 is arranged.

Next, the scale factor information extraction /
Details of the arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection / arrangement processing will be described with reference to the flowchart in FIG. In the bit allocation reading process in S38, the bit allocation arranged in the bit allocation information extracting / arranging process in S31 is read, and in S3
9 is output to the bit number '0' determination process.

The bit number '0' determination processing of S39 is for determining whether or not the bit allocation is '0'.
If it is “0”, the process proceeds to S15, and if it is not “0”, the process proceeds to S40. The scale factor extraction processing in S40 is for extracting a scale factor of audio encoded data of the MPEG audio standard. The scale factor selection / arrangement processing of S41 is for selecting and arranging the scale factor included in the encoded audio data of the MPEG audio standard.

The matrix operation coefficient selection processing in S42 is as follows.
As for the matrix calculation coefficients used in the band synthesis processing and the audio data output processing in S35, those for an effective band are selected from the matrix calculation coefficients defined in advance. The matrix operation coefficient arrangement processing in S43 is for arranging the matrix operation coefficients selected in the matrix operation coefficient selection processing in S42. The inverse quantization information selection processing of S44 is for selecting the predefined inverse quantization information. The inverse quantization information arrangement processing of S45 arranges the inverse quantization information selected in the inverse quantization information selection processing of S44.

Next, details of the quantization band division signal extraction / arrangement processing of S33 will be described with reference to the flowchart of FIG. In the bit allocation readout / quantization bit number analysis processing in S46, the bit allocation arranged in the bit allocation information extraction / arrangement processing in S31 is read and output to the bit number '0' determination processing in S39, and the allocated quantum is determined. It analyzes the number of coded bits.

The quantization band division signal extraction processing in S47 is as follows.
In accordance with the quantization bit number analyzed in the bit allocation readout / quantization bit number analysis processing of S46, the MPE
It extracts a quantized band division signal from audio encoded data of the G audio standard. The quantization band division signal arrangement processing in S48 arranges the quantization band division signals extracted in the quantization band division signal extraction processing in S47.

FIG. 2 shows details of the inverse quantization processing in S34.
This will be described with reference to the flowchart of FIG. The table limit number setting process in S49 includes bit allocation information extraction / placement process in S31, scale factor information extraction / placement / matrix operation coefficient selection / placement / inverse quantization information selection / placement process in S32, and quantization band division in S33. The table limit number is set based on the information arranged by the signal extraction / arrangement processing.

The band division signal arrangement processing in S50 is performed in S13.
The band division signal calculated by the scale factor multiplication processing is arranged. The table updating process in S51 is for updating the table currently being processed to the table to be processed next. The table limit number upper limit determination process of S52 is to determine whether or not the table limit number upper limit set in the table limit number upper limit setting process of S49 is reached. If it is not the upper limit, the processing after S09 is repeated, and if it becomes the upper limit, the inverse quantization processing ends.

Details of the band synthesizing process / audio data output process in S35 will be described with reference to the flowchart of FIG. The matrix operation process of S76 is to calculate the vectors V0 to V63 shifted by the shift process of S18 using the equation (1). Details of the matrix calculation processing S76 are shown in the flowchart of FIG.

Next, the operation of decoding encoded data in the second conventional example will be described. In FIG. 21, MPEG
When audio encoded data of the audio standard is input, the header information extraction / analysis processing of S02 extracts / analyzes the header information. Then, the bit allocation information extraction / placement processing of S31 extracts / places the bit allocation information. Based on the bit allocation information arranged here, the scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection / arrangement processing of S32 is performed by extracting / arrangement of scale factor information and matrix operation Selection / arrangement of coefficients and selection / arrangement of inverse quantization information are performed. However, selection and arrangement are not performed for a band having a bit number of “0”.

In the header information extraction / analysis processing in S02, the bit allocation information extraction / arrangement processing in S31, and the scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection / arrangement processing in S32, Information necessary for decoding is rearranged in a table.

Next, the quantization band division signal extraction /
The arrangement processing extracts a quantization band division signal from the audio encoded data. Also at this time, the scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement in S32 is performed.
As in the case of the inverse quantization information selection / placement processing, the selection / placement is not performed for the band having the bit number of “0”.

In the inverse quantization process in S34, the obtained quantized band division signal is inversely quantized to generate a band division signal. The difference between the inverse quantization process of S06 shown in FIG. 17 and the inverse quantization process of S34 shown in FIG. 21 is that in S06 the operation of the quantized band division signal is performed up to the set upper limit (all bands) of the number of bands. And calculating the quantization band division signal up to the upper limit (effective band) of the rearranged table.

Subsequently, the band division signal generated by the inverse quantization processing in S34 is synthesized with audio data by the band synthesis processing / audio data output processing in S35 and output. The difference between the band synthesis processing / audio data output processing of S07 shown in FIG. 17 and the band synthesis processing / audio data output processing of S35 shown in FIG. 21 is that the inverse quantization processing of S06 and the inverse quantization processing of S34 are different. Is similar to the difference.

In the band synthesizing process and the audio data output process in S35, the range of k in the expression (1) used in the matrix operation process in S76 is not 0 ≦ k <32 but 0 ≦ k <(valid (The number of bands = the number of tables limited). Thus, the encoded audio data of the MPEG audio standard is decoded into audio data.

[0053]

In the above-described conventional method for decoding encoded audio data, the upper limit of the frequency of the band division signal is determined only from the additional information in the encoded audio data. Therefore, even if there is a limitation on the reproducible frequency due to the reproducibility for the frequency of the audio replay device, such as a sound board attached to a computer (arithmetic processing device or image processing device), it is not taken into account and added. The audio encoded data has been decoded for all the frequencies set by the information. Therefore,
Since unnecessary band-divided signals are also decoded, the amount of calculation is increased and processing time is required.

In addition, it was not possible to limit the frequency of the band-divided signal other than the restriction by the additional information in the encoded audio data and the restriction in the time direction. Therefore, even if the processing capability of the arithmetic processing unit (hereinafter, referred to as a computer or a CPU) in the multitask operation is low, this is not taken into account, and audio encoded data is decoded for all band division signals set by the additional information. As a result, the amount of calculation increases and the processing time is increased. In the worst case, signal processing cannot be performed in time, and a state in which reproduction cannot be performed has occurred.

In addition, it is not possible to limit the frequency of the band-divided signal and the time direction by the setting of the user who performs the audio signal processing. For this reason, when audio encoded data is decoded on a computer, the load becomes heavy, and the computer may not be able to perform the processing (for example, image processing, document creation, etc.) originally intended.

The present invention has been made in view of the above-mentioned conventional problems, and makes it possible to actively limit the frequency of a band-divided signal and limit the time-direction band. High-speed, low-load audio-encoded data decoding by reducing the amount of computation optimally without the need to decode unnecessary band-division signals in accordance with the reproduction capability for different frequencies, the processing capability of the computer, and user settings It is an object to realize a method and an audio encoded data decoding device. It is another object of the present invention to provide a recording medium in which a program for such a signal processing is described.

[0057]

In order to solve such a problem, the invention according to claim 1 of the present application divides audio data into blocks each having a predetermined number of samples, and divides the audio data into a plurality of small bandwidths in units of blocks. Is converted into a band-divided signal, quantized for each band-divided signal, a signal obtained by combining the additional information is used as audio encoded data, and a signal obtained by converting the audio encoded data into a bit stream is input and input. An audio encoded data decoding method for decoding the bit stream into audio data, wherein the audio encoded data is extracted from an input bit stream, and additional information and a quantization band division signal are extracted from the audio encoded data. And when a playback capability notification signal indicating the frequency playback capability of the audio playback device is input, Discard the quantized band division signal of the band exceeding the reproduction capability of the device, perform inverse quantization on the quantization band division signal of the band not exceeding the reproduction capability of the audio reproduction device, and convert it to a band division signal. , By decoding the band-divided signal after the inverse quantization into audio data.

According to a second aspect of the present invention, audio data is divided into blocks each of a predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, and converted into band-divided signals. A signal obtained by converting the signal obtained by converting the audio encoded data into a bit stream and decoding the input bit stream into audio data; Encoding method, extracting the audio encoded data from an input bit stream, extracting additional information and a quantization band division signal from the audio encoded data, and quantizing each small bandwidth from the additional information. Bit allocation information indicating the bit amount allocation of the encoded band division signal, and from the additional information, Extracting scale factor information indicating the level of a bandwidth-quantized band-divided signal of a bandwidth, and when a reproduction capability notification signal indicating a frequency reproduction capability of an audio reproducing device is input, the quantization band of a band exceeding the frequency reproduction capability For the divided signal, the content of the bit allocation information is rewritten so that the number of reference locations is reduced, and for the quantized band divided signal of a band exceeding the frequency reproduction capability, the content of the scale factor information is reduced in the number of reference locations. Using the rewritten bit allocation information and the scale factor information, inverse quantization is performed on the quantized band division signal to convert it into a band division signal, and band division after inverse quantization is performed. It is characterized in that signals are band-synthesized and decoded into audio data.

Further, in the invention according to claim 3 of the present application, the reproduction capability notification signal can arbitrarily set its frequency reproduction capability.

Further, according to the invention of claim 4 of the present application, audio data is divided into blocks by a predetermined number of samples, band-divided into a plurality of small bandwidths in block units, and converted into band-divided signals. A signal obtained by converting the encoded signal into audio data, a signal obtained by converting the audio encoded data into a bit stream, and decoding the input bit stream into audio data An encoded data decoding method, wherein the audio encoded data is extracted from an input bit stream, additional information and a quantization band division signal are extracted from the audio encoded data, and each small bit is extracted from the additional information. Extract the scale factor information indicating the level of the quantized band division signal of the bandwidth,
When a processing capability notification signal indicating a processing speed in speech decoding of the arithmetic processing device at the present time is input, the content of the scale factor information is referred to for the quantized band division signal of a band exceeding the processing capability. Using the rewritten scale factor information, performs inverse quantization on the quantized band-divided signal, converts the quantized band-divided signal into a band-divided signal, and band-synthesizes the band-divided signal after inverse quantization. Thus, the audio data is decoded.

According to a fifth aspect of the present invention, the audio data is divided into blocks each having a predetermined number of samples, and the audio data is divided into a plurality of small bandwidths in units of blocks and converted into band-divided signals. A signal obtained by converting the encoded signal into audio data, a signal obtained by converting the audio encoded data into a bit stream, and decoding the input bit stream into audio data An encoded data decoding method, wherein the audio encoded data is extracted from an input bit stream, additional information and a quantization band division signal are extracted from the audio encoded data, and each small bit is extracted from the additional information. Extract the scale factor information indicating the level of the quantized band division signal of the bandwidth,
A decision is made as to whether or not the quantized band division signal having a value of the scale factor equal to or less than a certain value is discarded by thinning-out, and a processing capability notification signal indicating the current operation speed in speech decoding of the arithmetic processing device is input. When the quantization band division signal of the band exceeding the processing capacity, the value of the scale factor of the quantization band division signal determined to be discarded is set to zero, and the quantization band division signal that is not discarded The content of the scale factor information is rewritten so as to reduce the number of reference points, and the rewritten scale factor information is used to perform inverse quantization on the quantized band-divided signal and convert it to a band-divided signal. , By decoding the band-divided signal after the inverse quantization into audio data.

According to the invention of claim 6 of the present application, the processing capability notification signal can be arbitrarily set in accordance with the calculation speed of the processing unit.

According to a seventh aspect of the present invention, the audio data is divided into blocks each having a predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band divided signals, and converted into band divided signals. A signal obtained by converting the encoded signal into audio data, a signal obtained by converting the audio encoded data into a bit stream, and decoding the input bit stream into audio data An encoded data decoding method, wherein the audio encoded data is extracted from an input bit stream, additional information and a quantization band division signal are extracted from the audio encoded data, and each small bit is extracted from the additional information. Extracting bit allocation information indicating the bit amount allocation of the quantized band division signal of the bandwidth, The scale factor information indicating the level of the quantized band division signal of each small bandwidth is extracted from the additional information, a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproduction device, and the current operation speed in the audio decoding of the arithmetic processing device When a processing capability notification signal indicating the above is input, the content of the bit allocation information is rewritten so that the number of reference locations is reduced with respect to the quantized band division signal of a band exceeding the frequency reproduction capability and the arithmetic processing capability. When a processing capability notification signal indicating the operation speed in speech decoding of the arithmetic processing device is input, it is determined whether or not the quantized band division signal whose scale factor value is equal to or less than a certain value is discarded by thinning. Then, the value of the scale factor of the quantized band division signal determined to be discarded is set to zero, and Rewrite the content of the scale factor information with respect to the signal so as to reduce the number of reference points, and perform inverse quantization on the quantized band division signal using the rewritten bit allocation information and the scale factor information. , And convert the band-divided signal after dequantization into audio data by band-synthesizing the band-divided signal.

According to the invention of claim 8 of the present application, the audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band divided signals, and converted into band divided signals. A signal obtained by converting the encoded signal into audio data, a signal obtained by converting the audio encoded data into a bit stream, and decoding the input bit stream into audio data An encoded data decoding method, wherein the audio encoded data is extracted from an input bit stream, additional information and a quantization band division signal are extracted from the audio encoded data, and each small bit is extracted from the additional information. Extracting bit allocation information indicating the bit amount allocation of the quantized band division signal of the bandwidth, The scale factor information indicating the level of the quantized band division signal of each small bandwidth is extracted from the additional information, a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproduction device, and the current operation speed in the audio decoding of the arithmetic processing device When the processing capability notification signal indicating is input, for the quantized band division signal of the band exceeding the frequency reproduction capability and the arithmetic processing capability, the content of the bit allocation information is rewritten so that the number of reference locations is reduced, For the quantized band division signal of the band exceeding the frequency reproduction capability, the content of the scale factor information is rewritten so as to reduce the number of reference points, using the rewritten bit allocation information and the scale factor information,
Inverse decoding is performed on the quantized band-divided signal, the band-divided signal is converted into a band-divided signal, and the band-divided signal after the inverse quantization is band-synthesized to be decoded into audio data. .

According to the ninth aspect of the present invention, the audio data is divided into blocks each having a predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, and converted into band-divided signals. The signal obtained by quantizing the signal and the additional information is combined into audio encoded data, and the bit stream from the audio encoded data encoding device for converting the audio encoded data into a bit stream is input and decoded into audio data. An encoded data decoding device, comprising: audio encoded data decomposing means for extracting the audio encoded data from an input bit stream, and extracting additional information and a quantization band division signal from the audio encoded data. From the additional information extracted by the audio encoded data decomposing means, A bit allocation information extracting means for extracting bit allocation information indicating a bit amount allocation of the quantized band division signal of a small bandwidth; and a quantum of each small bandwidth from the additional information extracted by the audio coded data decomposing means. Scale factor information extracting means for extracting scale factor information indicating the level of the band-divided signal, a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproduction device, and a process indicating the current operation speed in the audio decoding of the arithmetic processing device When a capability notification signal is input, the number of reference locations for the contents of the bit allocation information extracted by the bit allocation information means is reduced with respect to the quantized band division signal having a band exceeding the frequency reproduction capability and the arithmetic processing capability. Rewritten as described above and extracted by the scale factor information extracting means. A capability analyzing means for instructing the content of the Kale factor information to be rewritten so that the number of reference points is reduced, and performing inverse quantization on the quantized band division signal rewritten based on the instruction of the capability analysis means, And a band-synthesizing audio data generating means for band-synthesizing the band-divided signal output from the inverse-quantizing means to decode the data into audio data. It is.

The invention according to claim 10 of the present application is characterized in that a program for the audio encoded data decoding method according to any one of claims 1 to 8 is described.

According to the first and second methods, the input audio coded data is divided, the reproduction capability for the frequency of the audio reproduction device is determined, and the quantization band division including the frequency not reproduced as audio is included. For signals,
Band synthesis is not performed.

According to the method of the third aspect, the user can arbitrarily set the reproduction capability with respect to the frequency.

According to the fourth aspect of the present invention, the arithmetic processing capability of the arithmetic processing unit is determined, the quantization of the quantized band-divided signal is performed according to the scale factor level, and the band of the band-divided signal after the inverse quantization is determined. I try not to synthesize.

According to the fifth aspect of the present invention, it is determined whether or not the quantized band division signal is to be thinned out, and the thinning is performed at regular intervals to reduce the amount of arithmetic processing involved in decoding.

According to the method of claim 6, the arithmetic processing capability of the arithmetic processing device can be set arbitrarily.

According to the seventh aspect of the present invention, the reproduction capability with respect to the frequency of the audio reproduction device and the arithmetic processing capability of the arithmetic processing device are determined, and the inverse quantization of the quantized band division signal including the frequency not reproduced as audio is determined. The band synthesis of the band-divided signal after the quantization and the inverse quantization is not performed.
Further, the quantization band division signal is thinned out at regular time intervals according to the scale factor level, thereby reducing the amount of arithmetic processing involved in decoding.

[0073]

BEST MODE FOR CARRYING OUT THE INVENTION

(Embodiment 1) Hereinafter, an audio encoded data decoding method according to Embodiment 1 of the present invention will be described with reference to the drawings. In the audio encoded data decoding method according to the present embodiment, a reproduction capability notification signal for notifying the reproduction capability with respect to the frequency of an audio reproduction device (hereinafter, referred to as a reproduction device) is input, and audio data that is not reproduced is decoded. It is characterized by not being performed. When a sound board is used as a playback device, its bit rate is determined by specifications. Therefore, compared to a device that reproduces a high-fidelity digital audio signal obtained by converting a normal analog audio signal into a PCM signal at a high sampling frequency,
It can be said that the reproducing apparatus as described above has a low reproducing ability. The reproduction capability notification signal described below is information for the reproduction device to notify the reproduction capability.

FIG. 1 is a flowchart showing an outline of signal processing in the audio coded data decoding method according to the present embodiment. In the flowchart of FIG.
The same processing symbols are assigned to the same signal processing as in the conventional example, and the description thereof is omitted. The input signal is audio encoded data of the MPEG audio standard.

In FIG. 1, the dynamic inverse quantization process in S53 performs inverse quantization on a quantized band division signal that can be reproduced by a reproduction apparatus based on a reproduction capability notification signal, and generates a band division signal. . The quantization band division signal is extracted in the quantization band division signal extraction processing of S05. A detailed flowchart of the dynamic inverse quantization process in S53 is shown in FIG. 2 described later.

The dynamic band synthesizing process / audio data output process of S54 is a process of synthesizing a band of a band-divided signal reproducible by a reproducing apparatus in accordance with a reproducing capability notification signal, converting the signal into audio data, and outputting the audio data. The detailed flow chart of the dynamic band synthesis processing / audio data output processing of S54 is shown in FIG. 3 described later.

The details of the dynamic inverse quantization processing in S53 will be described with reference to the flowchart in FIG. In the dynamic band number upper limit setting process of S55, if the band is not limited by the input reproduction capability notification signal, the header information extracted / analyzed in the header information extraction / analysis process of S02 shown in FIG. 1 is input. Then, the upper limit of the number of bands is set by the bit rate, sampling rate, and layer included in the header information. If the band is limited by the reproduction capability notification signal, the dynamic band number upper limit setting process of S55 sets the upper limit of the band number.

Next, the operation of decoding audio encoded data according to the present embodiment will be described. S shown in FIG.
When the quantization band division signal is extracted by the quantization band division signal extraction processing of 05, the dynamic inverse quantization processing of S53 performs inverse quantization to generate a band division signal. The dynamic inverse quantization processing in S53 performs signal processing shown in the flowchart of FIG.

In the dynamic band number upper limit setting process in S55, if there is no restriction by the reproduction capability notification signal, the bit rate included in the header information extracted and analyzed in the header information extraction / analysis process in S02 , The sampling rate, and the layer, the upper limit of the number of bands is set. If there is a limit due to the reproduction capability notification signal, the dynamic band number upper limit setting process of S55 sets the upper limit of the number of bands based on the bit rate, sampling rate, layer, and reproduction capability notification signal.

For example, it is assumed that the number of bands of 0 to 22.050 kHz is set to “32” from the bit rate, sampling rate, and layer. If the reproduction capability notification signal at this time is a signal meaning that "there is no limit on the frequency that can be reproduced by the reproduction device", the upper limit of the number of bands is set to "32" as it is. In addition, the reproduction capability notification signal indicates that the reproduction device can reproduce a frequency of 11.25 k
If it is a signal that means "to Hz."
Since 025 kHz is half the frequency of 22.050 kHz, the upper limit of the number of bands is half of '32', that is, '16'.
Is set to In this case, the processing from the inverse quantization information reading processing in S09 in FIG. 2 to the band updating processing in S15 is halved, and the amount of calculation is reduced.

The subsequent processing is the same as the conventional MPEG audio decoding method. Subsequently, audio data is synthesized and output from the band division signal generated by the dynamic inverse quantization processing in S53 by the dynamic band synthesis processing and audio data output processing in S54.

FIG. 3 is a flowchart showing the dynamic band synthesis processing and audio data output processing in S54. In the dynamic band number upper limit setting process in S55, the same process as the dynamic quantization upper limit setting process in S53 is performed to set the upper limit of the band number. Subsequent processing is the same as the conventional MPEG audio decoding method. However, the range of k in the equation (1) used in the matrix calculation process in S19 is not 0 ≦ k <32 but 0 ≦ k <(the number of band-limited bands). This reduces the amount of computation.

As described above, the contents of the reproduction capability notifying signal for notifying the reproduction capability with respect to the frequency of the reproduction device are determined for the audio encoded data of the MPEG audio standard. By not performing a part of the inverse quantization process and the band synthesis process of the audio encoded data that is not reproduced, the amount of calculation is reduced, and the encoded audio data can be decoded at a high speed with a small load. Also,
Although the playback capability notification signal is set and output by signal processing in the playback device or the decoding device, the user may arbitrarily set and output the playback capability notification signal.

FIG. 30 shows a band that cannot be reproduced and a band to be processed. FIG. 30 (A) shows a conventional method.
Processing bandwidth. On the other hand, FIG. 30B shows the method of the present embodiment, in which the processing band is set according to the reproduction band of the reproduction device.

(Embodiment 2) Next, an audio encoded data decoding method according to Embodiment 2 of the present invention will be described with reference to the drawings. FIG. 4 is a flowchart showing an outline of signal processing in the audio encoded data decoding method according to the present embodiment. As in the first embodiment, the audio coded data decoding method according to the present embodiment receives a reproduction capability notification signal for notifying the reproduction capability with respect to the frequency of the reproduction device, and decodes audio data that is not reproduced. It is characterized by not being performed.

Further, the audio coded data decoding method according to the present embodiment is described in the second prior art "Japanese Patent Laid-Open No. 8-25.
The audio encoding data decoding method used in the first embodiment is used for the decoding method of “Compressed audio information decompression method disclosed in Japanese Patent No. 1032”. here,
Compared with the first conventional MPEG audio standard decoding method, the second conventional decoding method relatively easily uses the audio coded data decoding method of the first embodiment. be able to. Here, from Embodiment 1,
Only the portions that are different from the decoding method of “Compressed audio information decompression method disclosed in Japanese Patent Application Laid-Open No. Hei 8-251032” are described.

Also in the present embodiment, the same portions as those of the signal processing described in the conventional example and the first embodiment are denoted by the same reference numerals, and description thereof will be omitted. In FIG. 4, the dynamic bit allocation information extraction / placement processing of S56 is performed by extracting the bit allocation information from the audio encoded data of the MPEG audio standard input by the audio encoded data input processing of S01 by the reproduction capability notification signal. Arrangement (rearrangement) is performed. A detailed flowchart of the dynamic bit allocation information extraction / placement processing of S56 is shown in FIG. 5 described later.

Extraction of dynamic scale factor information in S57 /
The arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing is performed by M
It extracts and selects and arranges scale factor information of audio encoded data of the PEG audio standard, selects and arranges matrix operation coefficients, and selects and arranges inverse quantization information. A detailed flowchart of the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S57 is shown in FIG.

The dynamic quantization band division signal extraction / arrangement processing of S58 is for extracting / arrangement of the quantization band division signal of the audio coded data of the MPEG audio standard based on the reproduction capability notification signal. A detailed flowchart of the dynamic quantization band division signal extraction / arrangement processing of S58 is shown in FIG. 7 described later.

The dynamic bit allocation information extraction / placement processing of S56 in FIG. 4 determines the content of the reproduction capability notification signal for notifying the reproduction capability with respect to the frequency of the reproduction device as shown in FIG. On the other hand, the signal is not decoded.

The dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S57 in FIG. 4 is performed as shown in FIG. The content of the reproduction capability notification signal for notifying the reproduction capability is determined, and the signal is not decoded for the audio data that is not reproduced.

In the dynamic quantization band division signal extraction / arrangement processing of S58 in FIG. 4, as shown in FIG. 7, the contents of the reproduction capability notification signal for notifying the reproduction capability with respect to the frequency of the reproduction device are determined, and the audio that is not reproduced is determined. For data, the signal is not decoded.

Next, the operation of decoding audio encoded data according to the present embodiment will be described. As in the decoding method of the second conventional example, the header information extraction / analysis processing of S02 shown in FIG. 4 extracts and analyzes the header information.
Next, in the dynamic bit allocation information extraction / placement processing of S56, the bit allocation information is extracted and placed according to the reproduction capability notification signal. At this time, if there is no band limitation by the playback capability notification signal, the dynamic band number upper limit setting process of S55 in FIG. 5 is based on the bit rate, sampling rate, and layer information analyzed from the header information. Set. If there is a band limitation due to the reproduction capability notification signal, the dynamic band number upper limit setting process of S55 sets the upper limit of the number of bands based on the bit rate, sampling rate, layer, and reproduction capability notification signal.

Extraction of dynamic scale factor information in S57 /
The arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing extracts scale factor information based on the bit allocation information and the reproduction capability notification signal rearranged as shown in FIG. / Relocation, selection / relocation of matrix operation coefficients, and selection / relocation of inverse quantization information. However, selection and rearrangement are not performed for a band in which the number of bits is '0' and a band exceeding the upper limit of the number of bands. Also, the header information extraction / analysis processing of S02, the dynamic bit allocation information extraction / arrangement processing of S56, the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement of S57 shown in FIG.
For dynamic inverse quantization information selection / placement processing, necessary information is rearranged in a table.

Then, based on the reproduction capability notification signal, S5
In the dynamic quantization band division signal extraction / placement processing of No. 8, the quantization band division signal is extracted from the audio encoded data.
At this time, the dynamic scale factor information extraction /
As in the case of the arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing, for the band in which the number of bits is '0' and the band exceeding the upper limit of the number of bands, Do not select or relocate.

In the inverse quantization process in S34, the quantized band division signal is inversely quantized to generate a band division signal. Subsequently, the band division signal generated by the inverse quantization processing in S34 is synthesized with audio data and output by the band synthesis processing / audio data output processing in S35.

In this way, when audio encoded data of the MPEG audio standard is input, the contents of the reproduction capability notifying signal for notifying the reproduction capability with respect to the frequency of the reproducing apparatus are determined. , Do not perform a part of the inverse quantization processing and the band synthesis processing. This makes it possible to perform high-speed and low-load audio encoded data decoding with a reduced amount of calculation. Although the playback capability notification signal is set and output by the signal processing of the playback device or the decoding device, the user may set and output the playback capability notification signal.

(Embodiment 3) Next, an audio encoded data decoding method according to Embodiment 3 of the present invention will be described with reference to the drawings. When a computer (CPU) performs a multitask operation, real-time processing capacity may be reduced. In this case, the CPU outputs a processing capacity notification signal for notifying the current processing capacity. The audio encoded data decoding method according to the present embodiment is characterized in that the content of a processing capability notification signal of a computer is determined, a scale factor level is set, and some audio data is not decoded. Is what you do.

FIG. 8 is a flowchart showing an outline of signal processing in the audio coded data decoding method of the present embodiment. Also in the present embodiment, the same portions as those of the signal processing described in the conventional example and the first and second embodiments are denoted by the same reference numerals, and description thereof will be omitted. In FIG.
S59 Dynamic scale factor information extraction / location / dynamic matrix operation coefficient selection / location / dynamic inverse quantization information selection /
In the arrangement processing, extraction / selection and rearrangement of scale factor information, selection / rearrangement of matrix operation coefficients, and selection / rearrangement of inverse quantization information are performed by a processing capability notification signal. FIG. 9 shows a detailed flowchart of the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S59.

In FIG. 9, in the scale factor comparing process in S60, a level for comparison with the scale factor is set by the processing capability notification signal, and the set level and the scale extracted in the scale factor extracting process in S40 are set. Compare the size with the factor level. If the level of the extracted scale factor is equal to or smaller than the set level (small), the process proceeds to S61. If the level of the scale factor exceeds the set level (large), the process proceeds to S41. Bit number '0' of S61
When the process proceeds to the setting process, the number of bits of the bit allocation is set to “0”. If the level of the extracted scale factor exceeds the set level, the scale factor selection / placement processing of S41 is executed.

Next, the operation of decoding audio encoded data according to the present embodiment will be described. As in the decoding method shown in the second conventional example, based on the bit allocation information rearranged by the bit allocation information extraction / arrangement processing of S31 in FIG. Scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing includes scale factor information extraction / arrangement, matrix operation coefficient selection / arrangement, and inverse quantization information Is selected / placed. However, selection and rearrangement are not performed for a band having a bit number of “0”.

Here, in the scale factor comparison process of S60 shown in FIG. 9, a level for comparison with the scale factor is set based on the processing capability notification signal. And S4
The level of the scale factor extracted in the zero scale factor extraction process is compared with the set level.

If the level of the extracted scale factor exceeds the set level, the process from the scale factor selection / placement process in S41 to the inverse quantization information placement process in S45 is executed. Then, the selection / arrangement of the extracted scale factor and the selection /
Processing of arrangement and selection / arrangement of inverse quantization information is performed. If the level of the extracted scale factor is equal to or less than the set level, the process proceeds to the bit number '0' setting process in S61,
The bit number of the bit allocation of the band of the extracted scale factor is set to '0'. In this case, S41
Without performing the processing from the scale factor selection / placement processing of step S45 to the inverse quantization information placement processing of step S45, the selection / placement of the extracted scale factors, the selection / placement of the matrix operation coefficients, and the processing of the inverse quantization information. Do not select / place.

By doing so, the amount of calculation is reduced. Also, by setting the number of bits to '0', the table set in the table limit number setting process in S49 in the inverse quantization process in S34 and the band synthesis process and audio data output process in S35 in FIG. The number limit is increased, the number of iterations is reduced, and the amount of computation is reduced.

As described above, when the audio coded data of the MPEG audio standard is input, the contents of the processing capability notification signal for notifying the processing capability of the computer are determined, and the scale factor level is set. For some of the audio encoded data, a part of the inverse quantization processing and the band synthesis processing is not performed. In this case, the amount of calculation is reduced, and high-speed and low-load decoding of audio encoded data can be performed. Further, the processing capability notification signal is described as being set and output by the computer, but the user may set and output the processing capability notification signal.

FIG. 31 shows the relationship between the scale factor set by the processing capacity and the level to be compared and the band to be processed. FIG. 31A shows a conventional method, in which the range of the band to be processed covers the entire frequency band even if the scale factor is equal to or less than the set level as indicated by the halftone dots. On the other hand, FIG. 31B shows the method according to the present embodiment, in which only the scale factor equal to or higher than the set level needs to be processed.

(Embodiment 4) Next, an audio encoded data decoding method according to Embodiment 4 of the present invention will be described with reference to the drawings. FIG. 10 is a flowchart showing an outline of signal processing in the audio encoded data decoding method according to the present embodiment. The audio encoded data decoding method according to the present embodiment determines the content of a processing capability notification signal for notifying the current arithmetic processing capability of the computer, sets the scale factor level, and sets a part of the audio data for a certain period of time. It is characterized in that decoding is not performed every time.

Also in this embodiment, the same portions as those of the signal processing described in the conventional example and the first, second, and third embodiments are denoted by the same reference numerals, and description thereof will be omitted. In FIG. 10, dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S62 is performed by extracting / arranging scale factor information by a processing capability notification signal. , Selection and arrangement of matrix operation coefficients and selection and arrangement of inverse quantization information. FIG. 11 shows a detailed flowchart of the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S62.

The dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing shown in FIG. 11 is performed by a processing capability notification signal for notifying the processing capability of the computer. The content is determined, the level of the scale factor is set, and some audio data is not decoded at regular time intervals.

In FIG. 11, in the status determination processing of S63, the status signal set in the status setting processing described later is determined. That is, if the status signal is set as “no time thinning”, the scale factor selection / arrangement processing of S41 is executed. If the status signal is set as “with time thinning”,
This is to execute the bit number '0' setting process of S61.

In the status setting process of S64, a status signal to be output to the status determination process of S63 is set based on the processing capability notification signal. The setting method here is realized, for example, by alternately repeating “with time thinning” and “without time thinning” every time the band is updated in the band updating process of S15.

Next, the operation of decoding audio encoded data according to the present embodiment will be described. Embodiment 3
As in the audio encoded data decoding method of
In the scale factor comparison process, the level of the scale factor extracted in the scale factor extraction process in S40 is compared with the level set based on the processing capability notification signal. If it exceeds, the processing from the scale factor selection / placement processing in S41 to the inverse quantization information placement processing in S45 is executed. Then, processing of selection / arrangement of the extracted scale factor, coefficient arrangement of matrix operation, and selection / arrangement of inverse quantization information is performed.

In the scale factor comparing process of S60, if the level of the extracted scale factor is equal to or less than the set level (small), the process proceeds to the status judging process of S63, and the status signal set in the status setting process S64. Is analyzed and determined. Here, if the status signal is set to "no time thinning", the processing from the scale factor selection / arrangement processing in S41 to the inverse quantization information arrangement processing in S45 is executed, and the extracted scale factor is selected. / Arrangement, matrix operation selection / arrangement, and inverse quantization information selection / arrangement.

In the status determination process of S63, if the status signal is set to "with time thinning out", the process proceeds to the bit number "0" setting process of S61, and the bit of the bit allocation of the band of the extracted scale factor is set. Set the number to '0'. In this case, the processing from the scale factor selection / arrangement processing of S41 to the inverse quantization information arrangement processing of S45 is not performed, and the selection / rearrangement of the extracted scale factors and the selection / rearrangement of the matrix operation coefficients are performed. , The selection / relocation of the inverse quantization information is not performed.
By doing so, the amount of calculation can be reduced stepwise as compared with the case of the third embodiment.

As described above, when the audio encoded data of the MPEG audio standard is input, the contents of the processing capability notification signal for notifying the processing capability of the computer are determined.
Set the scale factor level. Then, a part of the inverse quantization processing and the band synthesis processing is not performed on a part of the audio coded data at regular intervals. This makes it possible to reduce the amount of calculation step by step, and perform high-speed and low-load audio encoded data decoding.

Although the processing capacity notification signal has been described as being set and output by the computer, the user may set and output the processing capacity notification signal.

FIGS. 32 and 33 are explanatory diagrams showing the relationship between the scale factor set by the processing capacity, the comparison level, and the processing band. Here, as for the decimation at certain time intervals, “with time decimation” is performed every time the bandwidth is updated,
It is assumed that "no time thinning" is set to be repeated. FIG. 32 shows a conventional method, in which the processing band extends over the entire frequency band even if the scale factor is equal to or lower than the set level, as indicated by the halftone dots. On the other hand, FIG. 33 shows the method according to the present embodiment, and “no time thinning”.
Only the set scale factor of the band in the above need be processed. It should be noted that the band to be thinned out may have a scale factor value equal to or lower than a certain level, and the thinning band may be switched over time.

(Embodiment 5) Next, an audio encoded data decoding method according to Embodiment 5 of the present invention will be described with reference to the drawings. FIG. 12 is a flowchart showing an outline of signal processing in the audio encoded data decoding method according to the present embodiment. The audio encoded data decoding method according to the present embodiment uses the scale factor of the scale factor based on the content of the playback capability notification signal for notifying the playback capability for the frequency of the playback device and the content of the processing capability notification signal for notifying the processing capability of the computer. A level is set, and a part of audio data is not decoded at regular intervals.

Also in this embodiment, the same portions as those of the signal processing described in the conventional example and the first to fourth embodiments are denoted by the same reference numerals, and description thereof will be omitted.

In FIG. 12, the dynamic bit allocation information extraction / arrangement processing of S65 is for extracting and arranging bit allocation information based on a band control signal output by the capacity analysis / control processing of S68 described later. It is. FIG. 13 shows a detailed flowchart of the dynamic bit allocation information extraction / placement processing in S65.

The dynamic scale factor information extraction / placement / dynamic matrix operation coefficient selection / placement / dynamic inverse quantization information selection / placement processing of S66 in FIG.
Extracting / selecting and rearranging scale factor information based on the band control signal, the scale factor comparison control signal, and the status setting control signal output by the control process;
The selection / relocation of matrix operation coefficients and the selection / relocation of inverse quantization information are performed. Regarding the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S66,
FIG. 14 shows a detailed flowchart.

The dynamic quantization band division signal extraction / arrangement processing of S67 shown in FIG. 12 is based on the band control signal output by the performance analysis / control processing of S68, and is performed on the quantization band division signal of the audio encoded data. The extraction / arrangement is performed. A detailed flowchart of the dynamic quantization band division signal extraction / arrangement processing of S67 is shown in FIG. 15 described later.

The performance analysis / control processing of S68 in FIG.
Based on the reproduction capability notification signal and the processing capability notification signal, dynamic bit allocation information extraction / placement processing of S65 and S6
6 in the dynamic scale factor information extraction / placement / dynamic matrix operation coefficient selection / placement / dynamic inverse quantization information selection / placement processing and the dynamic quantization band division signal extraction / placement processing in S67. And a band control signal, a scale factor comparison control signal, and a status setting control signal.

The details of the dynamic bit allocation information extraction / placement processing in S65 will be described with reference to the flowchart in FIG. The dynamic band number upper limit setting process of S69 identifies the content of the input band control signal, and if the band is not limited, the header information extracted and analyzed in the header information extraction / analysis process of S02 is used. The upper limit of the number of bands is set by the included bit rate, sampling rate, and layer. If the band is limited, the dynamic band number upper limit setting process in S69 sets the upper limit of the number of bands based on the bit rate, sampling rate, layer, and band control signal.

Extraction of dynamic scale factor information in S66 /
Details of the arrangement / dynamic matrix calculation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing will be described. FIG.
In the scale factor comparison process in S70 of No. 4, a level for comparison with the scale factor is set based on the scale factor comparison control signal. Then, the set level is compared with the level of the scale factor extracted by the scale factor extraction processing in S40. If the level of the extracted scale factor is equal to or less than the set level, the status determination processing of S63 is executed.
If the level of the extracted scale factor exceeds the set level, the scale factor selection / placement process of S41 is executed.

The status setting process of S71 sets a status signal to be output to the status determination process of S63 based on the status setting control signal. The setting method is realized, for example, by alternately repeating “with time thinning” and “without time thinning” every time a band is updated in the band updating process in S15.

As shown in FIG. 15, the S described in FIG.
The dynamic quantization band division signal extraction / arrangement processing of 67 performs the scale factor level based on the reproduction capability notification signal notifying the reproduction capability for the frequency of the reproduction device and the processing capability notification signal notifying the processing capability of the computer. Is set so that some audio data is not decoded at regular time intervals.

Next, the operation of decoding audio encoded data according to the present embodiment will be described. First, FIG.
As shown in (11), the audio encoded data is input by the audio encoded data input processing of S01,
As a result of the performance analysis and control processing, a reproduction capability notification signal and a processing capability notification signal are input.

When the reproduction capability notification signal and the processing capability notification signal are input to the capability analysis / control process of S68, the capability analysis / control process includes the dynamic bit allocation information extraction / placement process of S65, The dynamic scale factor information extraction / placement / dynamic matrix operation coefficient selection / placement / dynamic inverse quantization information selection / placement processing of S66 and the dynamic quantization band division signal extraction / placement processing of S67 are optimal and It outputs a band control signal, a scale factor comparison control signal, and a status setting control signal so that signal processing can be performed at high speed.

Here, a method of outputting each of the band control signal, the scale factor comparison control signal, and the status setting control signal will be described. First, by the playback capability notification signal,
Outputs a band limited signal. At this point, if the amount of calculation that can be processed with the processing capacity notified by the processing capacity notification signal is equal to or less than the amount of calculation of decoding processing without decoding, only the band-limited signal is output. If the amount of calculation that can be processed with the processing capacity notified by the processing capacity notification signal greatly exceeds the amount of calculation of decoding processing without decoding, a scale factor comparison control signal is output following the band limitation signal. . Next, if the amount of operation that can be processed with the processing capability notified by the processing capability notification signal slightly exceeds the amount of operation of the decoding process without decoding, the scale factor comparison control signal Or a scale factor comparison control signal and a status setting control signal.

The header information extraction / analysis processing in S02 is as follows.
The header information is extracted from the encoded audio data and analyzed. Thereafter, the dynamic bit allocation information extraction / arrangement processing of S65 extracts / rearranges bit allocation information based on the band control signal. At this time, if there is no band limitation by the band control signal by the dynamic band number upper limit setting process of S69 shown in FIGS. 13, 14, and 15, the bit rate, sampling rate, and layer analyzed from the header information are Thus, the upper limit of the number of bands is set. If there is a band limitation by the band control signal, the upper limit of the number of bands is set by the bit rate, the sampling rate, the layer, and the band control signal.

Based on the bit allocation information, the band control signal, the scale factor comparison control signal, and the status setting control signal rearranged as described above, the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient of S66 is determined. Selection / disposition / dynamic inverse quantization information selection / disposition processing extracts, selects, and rearranges scale factor information, matrix operation coefficients, and inverse quantization information. However, selection and rearrangement are not performed for a band in which the number of bits is '0' and a band exceeding the upper limit of the number of bands.

Here, in the scale factor comparison process in S70 shown in FIG. 14, a level for comparison with the scale factor is set based on the scale factor comparison control signal, and the level is extracted in the scale factor extraction process in S40. The scale factor level is compared with the set level. If the level of the extracted scale factor exceeds the set level, the process from the scale factor selection / placement process in S41 to the inverse quantization information placement process in S45 is executed. Then, the selection / arrangement of the extracted scale factor and the selection /
Processing of arrangement and selection / arrangement of inverse quantization information is performed.

When the level of the extracted scale factor is
If the level is lower than the set level, the status signal set in the status setting process in S71 is analyzed in the status determination process in S63. If the status signal is set to "no time thinning", the processing from the scale factor selection / placement processing in S41 to the inverse quantization information placement processing in S45 is executed. Then, processing of selecting / arranging the extracted scale factor, selecting / arranging the matrix operation, and selecting / arranging the inverse quantization information is performed.

If the status signal is "time thinned out"
Is set, the number of bits in the bit allocation of the band of the extracted scale factor is set to '0' by the bit number '0' setting process in S61. Then, from the scale factor selection / arrangement processing of S41, S4
The processing up to the inverse quantization information arrangement processing of No. 5 is not executed. Also, selection / relocation of the extracted scale factor, selection / relocation of the matrix operation coefficient, and selection / relocation of the inverse quantization information are not performed.

Thereafter, based on the band control signal, S67
The dynamic quantization band division signal extraction / arrangement processing extracts a quantization band division signal from audio encoded data. At this time, as in the case of the dynamic scale factor information extraction / arrangement / dynamic matrix operation coefficient selection / arrangement / dynamic inverse quantization information selection / arrangement processing of S66, the number of bits is "0".
No selection or rearrangement is performed for the bandwidths set in the above and the bandwidths exceeding the upper limit of the number of bands.

The processing after the dynamic quantization band division signal extraction / arrangement processing of S67 is the same as that of the second conventional example.

As described above, when the audio encoded data of the MPEG audio standard is input, a reproduction capability notifying signal for notifying the reproduction capability with respect to the frequency of the reproduction device,
And the content of the processing capacity notification signal for notifying the processing capacity of the computer. Then, in order to set the level of the audio encoded data that is not reproduced and the scale factor, and to prevent decoding of some audio data at regular time intervals, the audio encoded data is subjected to inverse quantization processing and band synthesis processing. Do not do the department. In this case, the amount of calculation is optimally reduced, and high-speed, low-load audio encoded data decoding can be performed.

The playback capability notification signal is set in the signal processing of the playback device or the decoding device, and the processing capability notification signal is set and output by the computer. And a processing capability notification signal may be set and output.

(Embodiment 6) Next, an audio encoded data decoding apparatus according to Embodiment 6 of the present invention will be described with reference to the drawings. FIG. 16 is a block diagram illustrating a configuration of the audio coded data decoding device 1 according to the present embodiment. This audio encoded data decoding apparatus is an apparatus that implements the audio encoded data decoding method according to Embodiment 5 described above.

In FIG. 16, the audio coded data decoding device 1 comprises an audio coded data decomposing means 2, an inverse quantization means 3, a band synthesis audio data generating means 4, a capability analysis control means 5, a bit allocation information extracting means. 6. It includes a scale factor information extracting means 7.

The audio coded data decomposing means 2 is means for extracting audio coded data from the input bit stream and decomposing it into additional information and a quantization band division signal. The bit allocation information extracting means 6
This is a means for extracting bit allocation information indicating the bit amount allocation of the quantized band division signal of each small bandwidth from the additional information, and rearranging the bit allocation information based on the band control signal output from the capability analysis control means 5. .

The scale factor information extracting means 7 extracts scale factor information indicating the level of the quantized band division signal of each small bandwidth from the additional information, and
Means for rearranging the scale factor information based on the scale factor comparison control signal and the status setting control signal output by The rearranged bit allocation information and scale factor information are provided to the inverse quantization means 3.

The inverse quantization means 3 inversely quantizes the quantized band division signal based on the rearranged bit allocation information and scale factor information to generate a band division signal, similarly to the inverse quantization processing of S34. Is what you do. The band synthesizing audio data generating means 4 performs the band synthesizing process and the audio data generating process in S35, and synthesizes the band of the band-divided signal and outputs it as audio data.

The performance analysis control means 5 performs the performance analysis and control processing in S68, and can optimally and quickly process the audio encoded data decoding processing based on the reproduction performance notification signal and the processing performance notification signal. In this way, a band control signal, a scale factor comparison control signal, and a status setting control signal are output.

The operation of the audio coded data decoding device 1 having such a configuration will be described. First, audio encoded data is input to the audio encoded data decomposing means 2. At the same time, a reproduction capability notification signal and a processing capability notification signal are input to the capability analysis / control means 5. The bit allocation information extracting means 6 and the scale factor information extracting means 7 perform the band control signal and the scale factor comparison control signal so that the quantized band division signal can be inversely quantized by the inverse quantizing means 3 optimally and at high speed. , A status setting control signal, and outputs bit allocation information and scale factor information.

The audio encoded data decomposing means 2 extracts audio encoded data from the input bit stream, decomposes the encoded data into additional information and a quantization band division signal,
Extract matrix operation coefficients and inverse quantization information. The bit allocation information extracting means 6 extracts bit allocation information from the additional information obtained from the audio coded data decomposing means 2 and rearranges the information based on the band control signal. The scale factor information extracting means 7 extracts scale factor information of the quantized band division signal from the additional information, and based on the scale factor comparison control signal and the status setting control signal output from the capability analysis control means 5,
Rearrange the scale factor information.

The quantized band division signal is converted by the inverse quantization means 3 into a band division signal. This band division signal is synthesized with audio data by the band synthesis processing / audio generation means 4 and output.

(Embodiment 7) Next, a description will be given of a recording medium on which a program for a method for decoding encoded audio data according to Embodiment 7 of the present invention is recorded. The audio encoded data decoding method according to any of the first to fifth embodiments described above is realized by a program. Therefore, by recording these programs on a recording medium and transferring the programs, it is possible to easily carry out decoding of the encoded audio data by another independent computer.

[0150]

According to the audio encoded data decoding method according to the first and second aspects, the reproduction capability for the frequency of the audio reproducing apparatus is determined, and the inverse of the quantized band division signal including the frequency not reproduced as audio is determined. By not performing band synthesis of the band-divided signal after the quantization and the inverse quantization, the amount of calculation related to audio encoded data decoding can be reduced. Therefore, high-speed decoding with a small load can be realized.

According to the audio encoded data decoding method of the third aspect, in addition to the above effects, the user can determine the setting of the reproduction capability for the frequency.

According to the audio encoded data decoding method of the fourth aspect, the arithmetic processing capability of the arithmetic processing unit is determined, and the inverse quantization of the quantized band division signal and the dequantization of the band division signal after the inverse quantization are performed. By not performing band synthesis, the amount of calculation of the processing unit can be reduced. Therefore, high-speed and low-load decoding can be realized even when the arithmetic processing unit is executing another task.

According to the audio encoded data decoding method of the fifth aspect, the amount of calculation can be reduced in accordance with the processing capacity of the processing unit by thinning out the quantized band division signal at regular intervals. . Therefore, high-speed and low-load decoding can be realized even when the arithmetic processing unit is executing another task.

According to the audio coded data decoding method of the present invention, the user can determine the decoding processing capability of the arithmetic processing unit, thereby arbitrarily reducing the amount of calculation related to audio coded data decoding. Can be.

According to the audio coded data decoding method of the seventh aspect, by judging the reproduction capability for the frequency of the audio reproducing device and the processing capability of the arithmetic processing device, the quantum including the frequency that is not reproduced as the audio is determined. It is possible to reduce the processing of the inverse quantization of the band-divided signal and the band synthesis of the band-divided signal after the inverse quantization. In addition, by thinning out the band-divided signal at certain time intervals according to the scale factor level, the amount of calculation related to audio encoded data decoding can be significantly reduced.

According to the audio encoded data decoding method of the eighth aspect, in the audio encoded data decoding method of the seventh aspect, in the step of judging the ability, the reproduction ability setting for the frequency of the user is performed. Also, by judging the decoding processing capability setting for the computer of the user, there is an effect that the amount of calculation can be reduced optimally in accordance with the setting of the user, and a high-speed, low-load decoding method can be provided.

[Brief description of the drawings]

FIG. 1 is a flowchart showing main signal processing of an audio encoded data decoding method according to Embodiment 1 of the present invention.

FIG. 2 is a flowchart showing the contents of a dynamic inverse quantization process in the audio encoded data decoding method according to the first embodiment.

FIG. 3 is a flowchart showing the contents of a dynamic band synthesis process and an audio data output process in the audio encoded data decoding method according to the first embodiment.

FIG. 4 is a flowchart showing main signal processing of an audio encoded data decoding method according to Embodiment 2 of the present invention.

FIG. 5 is a flowchart showing the contents of dynamic bit allocation extraction / placement processing in the audio encoded data decoding method according to the second embodiment.

FIG. 6 shows a method of extracting / arranging dynamic scale factor information in the audio encoded data decoding method according to the second embodiment.
It is a flowchart which shows the content of dynamic matrix operation coefficient selection / placement / dynamic inverse quantization information selection / placement processing.

FIG. 7 is a flowchart showing the content of a dynamic quantization band division signal extraction / arrangement process in the audio encoded data decoding method according to the second embodiment.

FIG. 8 is a flowchart showing main signal processing of an audio encoded data decoding method according to Embodiment 3 of the present invention.

FIG. 9 shows dynamic scale factor information extraction / arrangement in the audio encoded data decoding method according to the third embodiment.
It is a flowchart which shows the content of dynamic matrix operation coefficient selection / placement / dynamic inverse quantization information selection / placement processing.

FIG. 10 is a flowchart showing main signal processing of an audio encoded data decoding method according to Embodiment 4 of the present invention.

FIG. 11 shows the contents of dynamic scale factor information extraction / allocation / dynamic matrix operation coefficient selection / allocation / dynamic inverse quantization information selection / allocation processing in the audio encoded data decoding method according to the fourth embodiment. It is a flowchart.

FIG. 12 is a flowchart illustrating main signal processing of an audio encoded data decoding method according to Embodiment 5 of the present invention.

FIG. 13 is a flowchart showing the contents of dynamic bit allocation extraction / placement processing in the audio encoded data decoding method according to the fifth embodiment.

FIG. 14 shows contents of dynamic scale factor information extraction / allocation / dynamic matrix operation coefficient selection / allocation / dynamic inverse quantization information selection / allocation processing in the audio encoded data decoding method according to the fifth embodiment. It is a flowchart.

FIG. 15 is a flowchart showing the contents of a dynamic quantization band division signal extraction / arrangement process in the audio encoded data decoding method according to the fifth embodiment.

FIG. 16 is a block diagram illustrating a configuration of an audio encoded data decoding device according to Embodiment 6 of the present invention.

FIG. 17 is a flowchart showing main signal processing in the first conventional encoded audio data decoding method.

FIG. 18 is a flowchart showing the contents of an inverse quantization process in the first conventional encoded audio data decoding method.

FIG. 19 is a flowchart showing the content of band synthesis processing and audio data output processing in the first conventional encoded audio data decoding method.

FIG. 20 is a flowchart showing the contents of matrix operation processing in the first conventional encoded audio data decoding method.

FIG. 21 is a flowchart showing main signal processing in a second conventional encoded audio data decoding method.

FIG. 22 is a flowchart showing the contents of bit allocation extraction / arrangement processing in a second conventional audio encoded data decoding method.

FIG. 23 is a flowchart showing contents of scale factor information extraction / arrangement / matrix operation coefficient selection / arrangement / inverse quantization information selection / arrangement processing in the audio encoded data decoding method of the second conventional example.

FIG. 24 is a flowchart showing the contents of a quantization band division signal extraction / arrangement process in a second conventional audio encoded data decoding method.

FIG. 25 is a flowchart showing the contents of an inverse quantization process in the audio encoded data decoding method of the second conventional example.

FIG. 26 is a flowchart showing the contents of band synthesis processing and audio data output processing in the audio encoded data decoding method of the second conventional example.

FIG. 27 is a flowchart showing the contents of a matrix calculation process in the audio encoded data decoding method of the second conventional example.

FIG. 28 is an explanatory diagram showing an arrangement of vectors V before and after a shift process in a decoding method according to the MPEG audio standard.

FIG. 29 is an explanatory diagram showing an arrangement of a vector V and a vector U before and after the vector arrangement in a decoding method according to the MPEG audio standard.

FIG. 30 is an explanatory diagram showing a relationship between a band that cannot be reproduced and a processing band in the audio encoded data decoding method according to the first embodiment.

FIG. 31 is an explanatory diagram showing a relationship among a set scale factor, a comparison level, and a processing band in the audio encoded data decoding method according to the third embodiment.

FIG. 32 is a diagram (part 1) illustrating a relationship between a set scale factor, a comparison level, and a processing band in the audio encoded data decoding method according to the fourth embodiment.

FIG. 33 is a diagram (part 2) illustrating a relationship between a set scale factor, a comparison level, and a processing band in the audio coded data decoding method according to the fourth embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 audio encoded data decoding device 2 audio encoded data decomposing means 3 inverse quantization means 4 band synthesis processing audio data generating means 5 capability analysis control means 6 bit allocation information extracting means 7 scale factor information extracting means

 ──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Hidenori Tatsumi 1-12-20 Hikaricho, Higashi-ku, Hiroshima City, Hiroshima Prefecture Inside Matsushita Electric Information System Hiroshima Research Institute Co., Ltd. (72) Inventor Eiji Kawahara Higashi-ku, Hiroshima City, Hiroshima Prefecture Matsushita Electric Information System Hiroshima Research Institute

Claims (10)

[Claims] An audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band division signals, quantized for each band division signal, and combined with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input, and an audio encoded data decoding method for decoding the input bit stream into audio data, comprising: Extracting the encoded audio data from the bitstream, extracting additional information and a quantized band division signal from the encoded audio data, and when a reproduction capability notification signal indicating a frequency reproduction capability of the audio reproduction device is input, Discarding the quantized band division signal of a band exceeding the reproduction capability of the audio reproduction device; By converting the band division signal by performing inverse quantization on the band quantized band division signal does not exceed the regenerative capacity of the raw device, band synthesizing band division signals after inverse quantization,
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 2. The audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and combined with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input, and an audio encoded data decoding method for decoding the input bit stream into audio data, comprising: Extracting the encoded audio data from the bit stream, extracting additional information and a quantized band division signal from the encoded audio data, and allocating a bit amount of the quantized band division signal of each small bandwidth from the additional information. Extracted bit allocation information, and quantized band division signals of each small bandwidth from the additional information. When a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproducing device is input, scale factor information indicating the level is extracted, and the bit allocation is performed on the quantized band division signal of a band exceeding the frequency reproduction capability. The content of the information is rewritten so that the number of reference locations is reduced, and the content of the scale factor information is rewritten so that the number of reference locations is reduced for the quantized band division signal of a band exceeding the frequency reproduction capability. Using the allocation information and the scale factor information, inverse quantization is performed on the quantized band division signal to convert it to a band division signal, and the band division is performed on the band division signal after the inverse quantization.
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 3. The audio-encoded data decoding method according to claim 1, wherein the reproduction capability notification signal can arbitrarily set its frequency reproduction capability. 4. The audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and combined with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input to an arithmetic processing device, and the audio encoded data decoding method for decoding the input bit stream into audio data. Extracting the audio encoded data from the input bit stream, extracting additional information and a quantization band division signal from the audio encoding data, and extracting a quantization band division signal of each small bandwidth from the additional information. The scale factor information indicating the level is extracted, and the performance in speech decoding of the arithmetic processing unit at the present time is extracted. When a processing capacity notification signal indicating a speed is input, the content of the scale factor information is rewritten to reduce the number of reference locations for the quantized band division signal of a band exceeding the processing capacity, and the scale is rewritten. By using the factor information, inverse quantization is performed on the quantized band division signal, converted into a band division signal, and the band division is performed on the band division signal after the inverse quantization.
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 5. The audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and synthesized with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input to an arithmetic processing device, and the audio encoded data decoding method for decoding the input bit stream into audio data. Extracting the audio encoded data from the input bit stream, extracting additional information and a quantization band division signal from the audio encoding data, and extracting a quantization band division signal of each small bandwidth from the additional information. Extracting scale factor information indicating a level, wherein the value of the scale factor is equal to or less than a certain value; Determine whether or not to discard the demultiplexed band division signal by thinning, and when a processing capability notification signal indicating a computation speed in speech decoding of the arithmetic processing device at the present time is input, the bandwidth of the band exceeding the processing capability is For the quantized subband signal, set the scale factor value of the quantized subband signal determined to be discarded to zero, and refer to the contents of the scale factor information for the quantized subband signal that is not discarded. Using the rewritten scale factor information, inverse quantization is performed on the quantized band division signal to convert it into a band division signal, and the band division signal after the inverse quantization is subjected to band synthesis. By doing
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 6. The audio encoded data decoding method according to claim 4, wherein the processing capability notification signal can be arbitrarily set in accordance with a calculation speed of a processing unit. 7. The audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and combined with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input to an arithmetic processing device, and the audio encoded data decoding method for decoding the input bit stream into audio data. Extracting the audio encoded data from the input bit stream, extracting additional information and a quantization band division signal from the audio encoding data, and extracting a quantization band division signal of each small bandwidth from the additional information. Extract bit allocation information indicating bit amount allocation, and calculate quantum of each small bandwidth from the additional information. The scale factor information indicating the level of the band division signal is extracted, and a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproduction device and a processing capability notification signal indicating the current operation speed in the audio decoding of the arithmetic processing device are input. At this time, the content of the bit allocation information is rewritten so as to reduce the number of reference locations with respect to the quantized band division signal having a band exceeding the frequency reproduction capability and the arithmetic processing capability, and the current arithmetic operation speed in the audio decoding of the arithmetic processing device. When the processing capability notification signal indicating the above is input, it is determined whether or not the quantized band division signal whose scale factor value is equal to or less than a certain value is discarded by thinning, and the quantization determined to be discarded is performed. The value of the scale factor of the band division signal is set to zero, and the scale factor of the quantized band division signal that is not discarded is reduced. Rewrite the contents of the actor information to reduce the number of reference points, perform inverse quantization on the quantized band division signal by using the rewritten bit allocation information and the scale factor information, and convert it to a band division signal. By performing band synthesis on the band-divided signal after inverse quantization,
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 8. Audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and combined with additional information. A signal as audio encoded data, a signal obtained by converting the audio encoded data into a bit stream is input to an arithmetic processing device, and the audio encoded data decoding method for decoding the input bit stream into audio data. Extracting the audio encoded data from the input bit stream, extracting additional information and a quantization band division signal from the audio encoding data, and extracting a quantization band division signal of each small bandwidth from the additional information. Extract bit allocation information indicating bit amount allocation, and calculate quantum of each small bandwidth from the additional information. The scale factor information indicating the level of the band division signal is extracted, and a reproduction capability notification signal indicating the frequency reproduction capability of the audio reproduction device and a processing capability notification signal indicating the current operation speed in the audio decoding of the arithmetic processing device are input. At this time, for the quantized band division signal of the band exceeding the frequency reproduction capability and the arithmetic processing capability, the content of the bit allocation information is rewritten so that the number of reference locations is reduced, and the quantization of the band exceeding the frequency reproduction capability is performed. The content of the scale factor information is rewritten with respect to the band division signal so that the number of reference points is reduced. Using the rewritten bit allocation information and the scale factor information, inverse quantization is performed on the quantized band division signal. Conversion into a band-divided signal, and band-synthesizing the band-divided signal after dequantization. More,
A method for decoding encoded audio data, wherein the encoded data is decoded into audio data. 9. The audio data is divided into blocks for each predetermined number of samples, divided into a plurality of small bandwidths in block units, converted into band-divided signals, quantized for each band-divided signal, and combined with additional information. A signal is audio encoded data, and the bit stream from the audio encoded data encoding device that converts the audio encoded data into a bit stream is input, and the audio encoded data decoding device decodes the audio data. An audio encoded data decomposing unit that extracts the encoded audio data from the input bit stream, and extracts additional information and a quantization band division signal from the encoded audio data; From the extracted additional information, a quantization band division signal of each small bandwidth A bit allocation information extracting means for extracting bit allocation information indicating bit amount allocation; and a scale factor indicating a level of a quantized band division signal of each small bandwidth from the additional information extracted by the audio coded data decomposing means. Scale factor information extracting means for extracting information; a reproduction capability notification signal indicating a frequency reproduction capability of the audio reproduction device; and a processing capability notification signal indicating a current operation speed in the audio decoding of the arithmetic processing device. The content of the bit allocation information extracted by the bit allocation information unit is rewritten so that the number of reference locations is reduced for the quantized band division signal of the band exceeding the frequency reproduction capability and the arithmetic processing capability, and the scale factor information extracting unit Refer to the contents of the extracted scale factor information. Capacity analyzing means for instructing rewriting so that the number of illuminated portions is reduced, and inverse quantization for performing inverse quantization on the quantized band-divided signal rewritten based on the instruction of the capacity analyzing means, and converting the signal into a band-divided signal. And a band-synthesizing audio data generating unit for decoding the band-divided signal output from the dequantizing unit into audio data by band-synthesizing the band-divided signal. . 10. A recording medium on which a program for the method for decoding audio encoded data according to claim 1 is described.