JP2002329372A - Audio information recording medium and reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a DVD audio data structure which is capable of confining a transmission rate within a specified standard and has specifics of high sound quality. SOLUTION: The recording medium is recorded with the data structure added with the head data including the first speech data string (sample) formed by digitalizing a portion of the channels of the speech signals of plural channels by a first sampling frequency and a first quantizing bit number, the second speech data string (sample) formed by digitalizing the other channels of the speech signals of the plural channels by a second sampling frequency and a second quantizing bit number and the timing data for synchronizing the first and second speech data strings (samples).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio information recording medium and a reproducing apparatus, and more particularly to a method for recording a digital audio signal on a high-density recording medium such as an optical disk, and the present invention is effectively applied to the reproducing apparatus. It is something.

[0002]

2. Description of the Related Art In recent years, high-density recording optical discs capable of recording a main video signal, a plurality of types of sub-video signals accompanying the main video signal, and audio signals of a plurality of channels have been developed. This high-density recording optical disc is
VD. Hereinafter, this technique is referred to as DVD video.

A technology called DVD audio is being developed by applying this DVD video technology. The DVD audio has been developed as a specialized audio technology, and aims at high sound quality.

[0004]

In the development of DVD audio, there is a demand that the standard be realized in a form as similar as possible to the standard of the audio data structure of DVD video.

[0005] Therefore, an object of the present invention is to provide a digital audio recording medium and a reproducing apparatus which realize a DVD audio standard having a high sound quality specification by utilizing a standard of an audio data structure in a DVD video as much as possible. .

[0006]

According to the present invention, in order to achieve the above object, data read and processed by using a computer control device has at least an audio object and management information, and the audio object has , A first channel group among a plurality of channels of audio signals is digitized at a first sampling frequency and a first quantization bit number to form a first audio data sequence (sample sequence); A second channel or a second channel group of the channel audio signals is digitized at a second sampling frequency and a second quantization bit number to form a second audio data sequence (sample sequence); The first and second sample strings are paired on a frame basis, are arranged in a frame, and the frame array is accommodated in a pack including a header to form a pack string. Made is and, the set of the pack is managed as a cell, is defined as the set of cells is managed as a program, the management information,
An audio title set information management table and a program chain information table necessary for reproducing the audio signals of the plurality of channels are provided. The professional program chain information table includes general program chain information and information on a plurality of programs. A program information table including a number of each entry cell and a cell reproduction information table indicating a reproduction position of the cell of each program are included, and the header includes a quantization of a sample of the first channel group. It is characterized by including information on a word length, information on a first audio sampling frequency, information on a quantized word length of samples of the second channel group, and information on a second audio sampling frequency.

[0007]

Embodiments of the present invention will be described below with reference to the drawings.

Before describing the present invention, a recording format of an audio signal defined in the DVD video standard will be described.

First, in the data recording method of the present invention,
An arrangement of data according to the linear PCM method will be described. For the linear PCM data, for example, 16 bits, 20 bits, and 24 bits are arbitrarily adopted as quantization bits. Further, audio modes include monaural, stereo, three-channel, four-channel, five-channel, six-channel, seven-channel, and eight-channel modes.

Now, it is assumed that there are audio signals of eight channels A to H. These are 48 kHz or 9
It is sampled at a sampling frequency of 6 KHz and quantized. The quantization bit will be described by taking 20 bits as an example.

FIG. 1A shows a state in which audio signals A to H of up to eight channels are sampled, respectively. It is assumed that each sample is quantized to, for example, 20 bits. 2 more
Each sample of 0 bits indicates a state of being divided into a main word and an extra word.

The main word of each channel is indicated by a capital letter An-Hn, and the extra word is indicated by a small letter an-hn. The suffix n (n
= 0, 1, 2, 3, 3,...) Indicates a sample order. Here, the main word is 16 bits, and the extra word is 4 bits.

The signal A is A0 a0, A1 a1, A2 a2
, A3 a3, A4 a4..., The signal B is B0 b0
, B1 b1, B2 b2, B3 b3, B4 b4..., The signal C is represented by C0 c0, C1 c1, C2 c2, C3 c
The signal H is represented by H0 h0, H1 h1 like 3, C4 c4.
, H2 h2, H3 h3, H4 h4...

Next, FIG. 1B shows the arrangement format of the above words in a sample sequence when the above words are recorded on a recording medium.

First, each sample data of 20 (= M) bits is divided into a main word of 16 (= m1) bits on the MSB side and an extra word of 4 (= m2) bits on the LSB side. Next, 0 (= 2
The n) th main word is arranged collectively. Next to this, the first (= 2n + 1) -th main word of each channel is arranged collectively. Next, the 0th (= 2n) th extra word of each channel is arranged collectively. Then, 1 (= 2n + 1) of each channel
The extra words are grouped together (where n
= 0, 1, 2,...).

Here, a group in which the main words of each channel are collected is defined as one main sample. A group of extra words of each channel is defined as one extra sample. FIG. 1B shows A0 to H0 (main sample S0), A1 to H1 (main sample S1), and a0 to h0 (extra sample e) of each channel.
0), a1 to h1 (extra sample e1),... These are referred to as four samples or two pairs of samples.

In such a format, when performing data reproduction processing with a simple model (for example, a model operating in a 16-bit mode), the main word of one of the channels, or each of the two channels in the case of stereo, is used. The reproduction process may be performed by handling only the main word. When the data reproduction process is performed by an upper model (for example, a model operating in the 20-bit mode), the reproduction process is performed by handling the main word and the corresponding extra word. Just do it.

FIG. 1C shows the arrangement state of each sample using the specific bit numbers of the main sample and the extra sample.

As described above, in the state of the quantized linear PCM code, the following can be achieved by dividing a 20-bit code into a 16-bit main word and a 4-bit extra word. It is. When a model operating in the 16-bit mode handles a sample array, unnecessary portions can be easily discarded by performing data processing in 8-bit units in the extra sample area. This is because two extra samples are 4 bits × 8 channels and 4 bits × 8 channels. This is because this data can be processed (discarded) continuously eight times in 8-bit units.

The features of this data array are not limited to this embodiment. In both cases where the number of channels is odd and the number of extra words is 8 bits, the total number of bits of two consecutive extra samples is an integral multiple of 8 bits. By executing the 8-bit consecutive discarding process n times according to, the extra sample can be skipped.

In the state shown in FIG. 1B, after that, modulation processing may be performed and recorded on a recording medium (on a track of an optical disk). However, when recording is performed together with other control information and video information, Is preferably recorded in a form that facilitates time management in order to facilitate data handling and synchronization. Therefore, the following framing, frame grouping, and packetization are performed.

FIG. 1D shows an audio frame sequence. That is, first, the unit of the data of the fixed reproduction time is (1/600 second), and this is set as one frame. In one frame, 80 or 160 samples are allocated. When the sampling frequency when the audio signal is sampled is 48 KHz, one sample is 1/4800 seconds, and the time of one frame is
(１／48000) × 80 samples = 1/600 second. When the sampling frequency is 96 KHz, one sample is 1/9600 seconds, and (1/96
000) × 160 samples = 1/600 second. Thus, one frame is made up of 80 samples or 160 samples.

FIG. 2 shows one frame and one GOF.
(Group of frames). One frame is data of 80 or 160 samples and is 1/600 second, and one GOF is composed of 20 frames. Then, one GOF corresponds to a period of (1/600) seconds × 20 = 1/30 seconds. That is, this is the television frame frequency. Such a sequence of GOFs is an audio stream. By deciding such a unit of 1 GOF, it is effective when synchronizing the audio stream and the video signal.

Further, the above-mentioned frames are allocated to packets for convenience of recording on the same recording medium as other control signals and video signals. The relationship between the packet and the frame will be described below.

FIG. 3A shows the relationship between the packet and the frame.

NV is a navigation pack, which includes a pack header, a packet header, a PCI_PKT
(Presentation control packet), and D
SI_PKT (data search information packet) is described. The DSI_PKT data is data search information, where V is a pack of video objects, A is a pack of audio objects, and S is a pack of sub-picture objects. One pack is defined as 2048 bytes. One pack includes one packet, and one pack includes a pack header, a packet header, and a packet. DSI_P
In the KT data, information for controlling each data at the time of reproduction, such as a start address and an end address of each pack, is described.

FIG. 3B shows only the audio pack taken out. Actually, DSI_PKT, video pack V, and audio pack A are mixedly arranged as shown in FIG. 3A, but FIG. 3B makes it easy to understand the relationship between frames and packs. for,
Audio pack A is taken out and shown. In the standard of this system, DSI_PKT and the next DSI_PKT
It is stipulated that information that is only about 0.5 second when the data is reproduced between and is arranged. Therefore, since one frame is 1/600 second as described above, DSI_
The number of audio frames between the PKT and the next DSI_PKT is 30 frames. The data amount (D) of one frame differs depending on the sampling frequency (fs), the number of channels (N), and the number of quantization bits (m).

When fs = 48 kHz, D = 80 × N × m, and when fs = 96 kHz, D = 160 × N × m.

Therefore, one frame does not always correspond to one pack, and a plurality of frames may correspond to one pack, or one frame or less may correspond to one pack. That is, as shown in FIG. 3B, the beginning of the frame may come in the middle of one pack. The position information of the head of the frame is described in the pack header, and is described as the number of data counts (timing) from the pack header or DSI_PKT. Therefore, when reproducing the above-described recording medium, the reproducing device extracts the frame of the audio packet, extracts the data of the channel to be reproduced, takes it into the audio decoder, and performs the decoding process.

FIG. 4A shows the relationship between the main word (16 bits) and the extra word (4 bits) in the 20-bit mode, which generally shows the above data arrangement. FIG. The relationship between the main word (16 bits) and the extra word (8 bits) in the 24-bit mode is shown.

As shown in FIGS. 4A and 4B, the sample data is composed of a main sample and an extra sample as a pair, and two pairs of samples as one unit. Done.

As described above, it is possible to obtain an arrangement method, a medium, and a processing apparatus for recording or transmitting data of multi-channel linear PCM system data that can be reproduced by both a simple model and a high-order model. Can be.

In the standard of this system, DSI_PKT
It is stipulated that an information amount of about 0.5 second is arranged when information between the DSI_PKT and the next is reproduced.

One pack includes a pack header, a packet header, and a packet data part. The pack header and packet header contain the size of the audio pack,
Information necessary for audio reproduction such as presentation time stamp for obtaining reproduction output timing with video, channel (stream) identification code, quantization bit, sampling frequency, data start address, end address, etc. is described. ing. The audio inserted in the packet is shown in FIG.
(C) is inserted in units of two pairs of samples consisting of two main samples and two extra samples.

FIG. 5 shows an enlarged audio pack. In the data portion of this audio pack, the head of the pair of samples (A0-H0,
A1-H1), and are subsequently arranged in units of two pairs. Here, the number of bytes in one pack is fixed at 2048 bytes. On the other hand, since the sample is variable-length data, 2048 bytes is not necessarily a byte length that is an integral multiple of two pairs of samples. Therefore, the maximum byte length of one pack may be different from the byte length of (2 pairs of samples × an integer). In such a case, the byte length of the pack ≧ (two pairs × an integer multiple) is satisfied, and the following countermeasures are taken when a part of the pack remains. That is, if the remaining portion of the pack is 7 bytes or less, a stuffing byte is inserted in the pack header, and if the remaining portion exceeds 7 bytes, a padding packet is inserted at the end of the pack (shaded in FIG. 5). See the section that did.)

In the case of such packed audio information, handling is easy during reproduction.

Since the head audio data of each pack is always the head of two pairs of samples, that is, the main sample, the reproduction process is facilitated when the reproduction process is performed at a proper timing. This is because the playback device fetches data in packs and performs data processing. If the audio data samples are arranged over two packs, the two packs are fetched, the audio data is integrated and decoded, and the processing becomes complicated. However, as in this method, the audio data at the head of each pack is always the head of two pairs of main samples, and if audio data is grouped in packs, the timing is taken only for one pack. And the processing is easy. Further, since the data processing is performed in units of packets, the authoring system (support system) can be simplified, and the software for data processing can be simplified.

In particular, at the time of special reproduction or the like, video data may be intermittently thinned out for processing or interpolated for processing. In such a case, audio data can be handled in packet units. As a result, it is possible to relatively easily control the reproduction timing. It does not complicate the decoder software.

In the above system, the sample is created in such a manner that the sample is divided into upper 16 bits and lower 4 bits, but it is not always necessary to use such a format. What is necessary is just to sample linear PCM audio data.

For example, if the data length of the extra sample is 0
Considering the above, the data sequence is a sequence of main samples, and has a general data format. In this case, since there is no extra sample, it is not necessary to use two pairs of samples, and packetization may be performed in units of main samples.

FIG. 6 shows a list of linear PCM data sizes in a case where linear PCM data is arranged in a packet in units of two pairs as described above. The number of monaural, stereo, and multi-channel sections is shown, and in each section, the maximum number of samples that can be accommodated in one packet is shown for each quantization bit number. Since the number of samples is two, the number of samples in one packet is all even. As the number of channels increases, the number of bytes increases accordingly, so the number of samples in one packet decreases. When the quantization bit number is 16 bits and monaural, the number of samples in one packet is 1004, the number of bytes is 2008, the stuffing byte is 5 bytes, and there is no padding byte. However, the stuffing byte of the first packet is 2 bytes. This is because 3-byte attribute information may be added to the header of the first packet.

Looking at the 24-bit quantization bits and the stereo mode, it is shown that the leading packet is stuffed with 6 bytes and the subsequent packets are padded with 9 bytes.

FIG. 7 shows an outline of a pack header of an audio pack.

First, a pack start code (4 bytes)
And then the system clock reference (SCR)
Is described. System clock reference (S
(CR) indicates the loading time of this pack. If the value of this SCR is smaller than the value of the reference time inside the apparatus, the pack to which this SCR is added is loaded into the audio buffer. In the pack header, the program multiplexing rate is described in 3 bytes. Further, the stuffing length is also described in one byte.
The control circuit can determine the read address of the control information by referring to the stuffing length by the control circuit.

FIG. 8 shows the contents of the packet header of the audio packet. The packet header includes a packet start code prefix for notifying the start of the packet, a stream ID indicating what data the packet has, and data indicating the length of the packet stream. Various types of information of the packet elementary stream (PES), such as a flag indicating prohibition and permission of copying, a flag indicating original information or copied information, and a length of a packet header are described. Furthermore, a presentation time stamp (PTS) for synchronizing the output of the packet with another video or sub-picture is described. Further, in the first packet of the first field in each video object, information such as a flag indicating whether or not a buffer is described and the size of the buffer are described. It also has 0-7 byte stuffing bytes.

Further, it has an audio stream, a linear PCM or another compression system, and a substream ID for indicating the number of the audio stream.
Furthermore, the number of audio frames in which the first byte data is arranged in this packet is described.
Furthermore, a pointer indicating the first audio frame in the packet to be reproduced at the time pointed by the PTS, that is, the first byte of the unit to be accessed first is described. This pointer is described by the byte number from the last byte of this information. The pointer indicates the first byte address of the audio frame. Also, an audio enhancement flag indicating whether or not high-frequency enhancement has been performed, a mute flag for obtaining mute when audio frame data is all 0, and an audio frame group (GO)
The frame number to be accessed first in F) is also described. Also, the length of the quantization word, that is, the number of quantization bits, the sampling frequency, the number of channels, control information of the dynamic range, and the like are described.

The above header information is analyzed by a decoder control unit (not shown) in the audio decoder.
The decoder control unit switches the signal processing circuit of the decoder to a signal processing mode corresponding to the audio data currently being captured.

Since the same information as the above header information is also described in the video manager, if such information is read at the beginning of the reproducing operation, it is not necessary to read the same sub stream thereafter. Absent. However, as described above, the mode information necessary to reproduce audio is described in the header of each packet because, for example, when a packet sequence is transmitted in a communication sequence, reception is started at any time. This is because the receiving terminal can recognize the audio mode. Also, this is because audio information can be reproduced even when only the pack is taken in by the audio decoder.

In the above DVD video, the maximum transfer rate of audio data is 6.144 Mbps, and the maximum transfer rate of the total of all audio data streams is 9.8 Mbps. The attributes (sampling frequency fs, number of quantization bits Qb, number of channels, etc.) of each channel in one stream are the same. This restriction is defined in the DVD video standard.

Due to such restrictions, multi-channel audio such as surround (eg, six channels of R, L, C, SL, and SW exist in one stream) has a high sound quality specification. Can not be realized.

In other words, the sampling frequency fs and the number of quantization bits Qb of all the channels must be the same under the above constraint conditions, so that high sound quality (for example, fs = 96k)
Hz), it is necessary to cope with all channels in the same manner, so that the value of the transfer rate becomes large and exceeds the specified value.

For example, the transfer rate per channel (ch) at the sampling frequency fs and the number of quantization bits Qb is 2.304 Mbps / 96 kHz, 24 bits, purely for the audio data portion.
When the frequency is 96 kHz and 20 bits, 1.92 Mbps / c
h At 96 kHz, 16 bits, 1.536 Mbps /
When the channel is 48 kHz and 24 bits, the frequency is 1.152 Mbps /
0.948 Mbps / c for 20 kHz with 48 kHz ch
h 48 kHz, 16 bits, 0.76 Mbps / c
h, the high sound quality specification that can be realized under the above constraint conditions of the DVD video standard is 48 kHz, 20 bits, 6 bits.
Up to channels (the transfer rate for audio in this case is 0.96 × 6 = 5.76 Mbps <6.144)
, And further specifications are not supported.

Therefore, in the present invention, the data structure of the DVD audio standard having a high-quality audio signal specification is devised while keeping the audio data structure type in the DVD video standard as much as possible.

Hereinafter, the basic concept of the present invention will be described by comparing the DVD video standard and the DVD audio standard.

The size of the audio pack in DVD audio is set to 2048 bytes as in the case of DVD video. Also, the number of quantization bits is Qb = 16 bits or 20 bits in the same manner as the audio specification in DVD video.
it or 24 bits. However, in the case of DVD audio, the number of simultaneously transferred linear PCM audio streams is limited to one. That is, in the case of DVD video, when movie content is recorded as a video object, various languages are assigned to each channel of the audio stream, and stream switching can be selected. However, since DVD audio basically targets music content, it is not always necessary to make a switching selection for each stream, so that all channels can be used to reproduce and output simultaneously. Can be realized. In the system of the present invention, the linear PCM audio streams to be transferred simultaneously are unified into one.

Next, the maximum transfer rate for DVD audio was increased from 6.144 Mbps to 9.6 Mbps. As described above, when looking at the entire data stream of a DVD video, packs of video data, sub-picture data, audio data, navigation data, and the like are transmitted in a time-division multiplex manner. The maximum transfer rate including the entire transmission data is 9.6 Mbp.
s. For this reason, it is difficult to increase the transfer rate of audio data to 6.144 Mbps or more. However, compared to DVD video, DVD audio is all audio data except for some control data, so that the amount of audio data can be increased and the transfer rate can be increased.

As described above, the maximum transfer rate of DVD audio has been increased, so that the maximum transfer rate as shown in FIG.
The number of samples in an audio frame is reduced to half that of DVD video. Therefore, the sampling frequency fs
For fs = 48 kHz or 44.1 kHz, 40 samples / frame fs = 96 kHz or 88.2 kHz for 80 samples / frame fs = 192 kHz or 176.4 kHz for 160
Pieces / frame. (In the case of DVD video, 44.1 kHz, 8
8.2 kHz, 176.4 kHz and 192 kHz are not supported. This means that at least one audio pack is included in one audio frame, and the audio frame must be a presentation time stamp (PT
S) This is for having data (data for synchronizing with the system time stamp at the time of reproduction).

Here, in DVD audio,
A scalable method is used to achieve high-quality sound specifications that surpass DVD video. That is, while all channels in one stream have the same attribute with respect to the sampling frequencies fs and Qb, channels having different attributes are allowed in one stream.

This corresponds to, for example, R (right channel), L
(Left channel), surround C (center channel), SR (rear right channel), SL (rear left channel), and SW (low frequency channel), all channels having high sound quality (high sampling frequency fs). ), The main channels (for example, R and L) have high sound quality (for example, fs = 96 kHz), and the other sub channels (C, SR, SL, and S)
Even if W) is the current sound quality (fs = 48 kHz), it is based on the fact that the sound quality can be made sufficiently high as a whole.

Here, the concept of the audio system using the scalable method will be briefly described as follows.

The maximum transfer (transmission) rate of one channel group signal for audio is 6.144 Mbp.
The maximum transfer rate, which is the sum of the transfer rates of one stream of signals below s, is targeted to be 9.8 Mbps or less. The channel group is, for example, stereo R, L
It is a digital signal containing channels (main two channels). A stream composed of C, SR, SL, and SW is also one channel group.

Next, a case where an audio signal of, for example, 6 channels is recorded as a signal to be recorded on a recording medium will be described. The six channels mentioned here are, for example, R, L, C, SR, SL,
SW, and a signal corresponding to each channel is generated. It is also possible to distinguish R and L as main channels and the others as sub-channels. Then, if the signal of each channel is reproduced and supplied to the speaker, a three-dimensional sound effect is obtained. Here, according to the method of the present invention, the above-mentioned six channels are assigned to the first channel.
And a second channel group.
In this case, R and L with high importance are selected as channels constituting the first channel group, and C, SR, SL and SW are selected as channels constituting the second channel group. Here, the signal of the first channel group has a high sampling frequency fs, and the signal of the second channel group has a sampling frequency of fs / 2 (1 / integer).

FIG. 9 specifically shows a signal processing system of the first channel group and a signal processing system of the second channel group.

In the analog signal source 10, signals of R, L, C, SR, SL, and SW channels in the surround system are prepared and supplied to the sampling unit 11. The sampling section 11 samples the signal of each channel at 96 kHz, and the sampled signal is input to the quantization section 12, quantized to 24 bits, and converted into a PCM (pulse code modulation) signal.

Next, each of the signals of the C, SR, SL, and SW channels has a sampling frequency of 96 kHz, which is 1 /
2 is converted to a frequency of 48 kHz. 96 kHz R,
The signal of the L channel is input to the phase matching unit 14,
The phases are adjusted so that the phase correspondence between the samples can be obtained. Actually, the same delay amount as that of the frequency conversion unit 13 is set in the phase matching unit 14. Next, the delayed 96 kHz R and L channel signals are input to the framing unit 15 and are framed every predetermined number of samples.

Also, 48 kHz C, S after frequency number conversion
The signals of the R, SL, and SW channels are transmitted to the framing unit 1
6 and is framed every predetermined number of samples. Each of the framed signals is input to the packetizing unit 17 and is converted into a packet of a predetermined format. Then, a 96 kHz stream (first attribute At)
An r1 stream) and a 48 kHz stream (a stream with the second attribute Atr2) are obtained. However, these two streams have an identifier (I
D) is attached. The packets of the two channel groups are further packed, multiplexed and output, and recorded on the disk 18 via a recording processing unit (not shown).

When the signal recorded on the disk 18 is reproduced, the following processing is performed.

A signal optically read from the disk 18 is demodulated (not shown) for performing error correction, demodulation, and the like.
Is input to the packet processing unit 21 via the. The packet processing unit 21 identifies a channel group with reference to the identifier of the packet header. By this identification, the packets of the first channel group and the packets of the second channel group can be identified, and the signals of each channel group are sorted. So-called demultiplexing. The signal of the first channel group is input to the frame processing unit 22, where the frame is released, and output as R and L channel signals. The signal of the second channel group is input to the frame processing unit 23, the frame is released, and the signal is output as each signal of the C, SR, SL, and SW channels.

Here, the signals of the R and L channels are input to the phase matching unit 24, and the signals of the C, SR, SL and SW channels are used to convert (upconvert) the sample frequency from 48 kHz to 96 kHz. The signal is input to the frequency converter 25.

The signals of the R and L channels which are phase-matched and have the same sample frequency, and C, SR, SL,
Each signal of the SW channel is input to the 96 kHz digital-to-analog converter 26, PCM-decoded, converted to an analog signal, and output.

With the above processing, high-quality R and L channel signals and C, SR, SL and SW channel signals can be reproduced.

According to the present invention, as described above, when the sample data in one frame is reproduced,
The number of samples is set to be seconds. For this purpose, a 96 kHz stream (first channel group)
And a 48 kHz stream (second channel group)
Are different in the number of samples in one frame. FIG. 10 shows the samples existing in the frame in comparison between the first channel group and the second channel group. The phase matching unit 14 creates a frame by performing phase matching between the first channel group and the second channel group.

Then, the framing sections 15 and 16 add the head of the corresponding frame (the frame to be reproduced at the same time in time) of the first and second channel groups to
The same presentation time stamp is added. As a result, at the time of reproduction, the frame processing unit 22,
When the frame is released at 23 and supplied to the digital-to-analog converter, the release timing of each frame is
Frames having the same presentation time stamp may be simultaneously released.

As described above, in DVD audio,
Originally, the channel group forming one audio stream can be divided into two attribute groups Atr1 and Atr2. Here, the attributes include the sampling frequency fs, the number of quantization bits Qb, and the number of channels. Of course, if the attributes of all the channels in one stream are the same, they need not be divided into two attribute groups.

The arrangement of the case of six surround channels as in the above example is as follows.

As the attributes (Atr1) of the channel groups R and L, fs = 96 kHz, Qb = 24 bits, and the channel groups C, SR, SL, and SW attributes (Atr2)
There are two types, s = 48 kHz and Qb = 24 bits. Then, the transfer rate in this case is 2.30
4 × 2 + 1.1152 × 4 = 9.216 Mbps, which satisfies the above-described maximum transfer rate of 9.8 Mbps. Therefore, by introducing the scalable method, an audio data structure having a high-quality sound signal specification can be obtained.

In the above description, the first channel group, the second channel group
Sampling frequency f as an attribute in the channel group
s and the number of quantization bits Qb were considered.

In the method of the present invention, the sampling frequency is different for each channel group as an attribute of each channel group. When the number of quantization bits is the same, when the sampling frequency is the same and the number of quantization bits is different, the sampling frequency is the same.
In various combinations such as when the number of quantization bits is the same, when the sampling frequency is different, and when the number of quantization bits is different, the point is that the above-mentioned maximum transfer rate is 9.8 Mbp
A stream that satisfies s may be configured.

FIG. 11 shows case 1, in which attributes Art1, Art in the first channel group and the second channel group are used.
2 and the sampling frequency fs = 96 kHz and fs =
48 kHz is shown.

FIG. 12 shows the case 2.
In this case, as the attribute Atr1 of the first channel group,
fs = 96 kHz, attribute of the second channel group (Atr
2) shows a case where fs = 48 kHz.

FIG. 13 shows the case 3.
In this case, as the attribute Atr1 of the first channel group,
The case where fs = 48 kHz is shown as fs = 96 kHz and the attribute Atr2 of the second channel group.

When a channel group having different attributes exists in one stream as described above, the data structure of the present invention has the following data structure.

The data structure of FIG. 14 corresponds to the case 1 of FIG.
And the attribute Ar in the first channel group
As t1, a sampling frequency fs = 96 kHz and a quantization bit number Qs = 16 bits are adopted. As an attribute Art2 in the second channel group, fs = 48 kHz,
This is an example in which the number of quantization bits Qs = 16 bits is adopted.
This data structure is similar to the scalable method described above, and also has a data structure similar to the DVD video sample array structure.

That is, four samples S4n, S4n + 1, S4n + 2,
S4n + 4 is a main sample having the first attribute, and two samples S2n and S2n + 1 are main samples having the second attribute.
In this case, there are no extra samples because the number of quantization bits is 16 bits.

This example relates to the sampling frequency,
Two samples of the second channel group correspond to four samples of the first channel group. The first group of channels, which is the main, is basically composed of four samples, and if the second group of channels is added, the total number of samples is basically six.

That is, in the data structure of FIG. 14, the signals of the first channel group, which is at least two of the plurality of channels, are sampled at the first frequency,
The signal of the second channel group, which is another channel, is
S4n-th, S4n + 1-th, and S4n-th samples of each channel of the first channel group sampled at the first frequency.
4n + 2nd and S4n + 3th are sequentially arranged, and then S2nth and S2 of the samples of each channel of the second channel device sampled at the second frequency.
The (n + 1) th is sequentially arranged (however, n = 0, 1,
2, ...).

The data structure of FIG. 15 is the same as the case 2 of FIG.
And the attribute Ar in the first channel group
The sampling frequency fs = 96 kHz and the number of quantization bits Qs = 24 bits are adopted as t1, and the attributes Art2 in the second channel group are fs = 96 kHz,
This is an example in which the number of quantization bits Qs = 20 bits is adopted.
In this case, two pairs of samples S2n, S2n + 1, e2n, e2n
+1 includes a main sample and an extra sample of the first attribute, and other two pairs of samples S2n, S2n + 1, e2n, e2n
+1 is the main sample of the second attribute, and the whole is basically a four-pair sample. E2n and e2n + 1 of the first attribute are the second
There is an extra sample of attributes.

That is, in the data structure of FIG. 15, the signal of the first channel group, which is at least two of the plurality of channels, is sampled at the first frequency, and the signal of the second channel group, which is another channel, is sampled. The sampled data is sampled at the second frequency, and the sample data is further divided into the m1 bit main word on the MSB side and the LSB.
M2 bit extra words.

Then, the main word of the (2n) -th sample data of each channel of the first channel group is put together and arranged as a main sample S2n, and then the (2n + 1) -th of each channel of the first channel group is arranged. Are arranged as a main sample S2n + 1, and then the extra words of the (2n) -th sample data of each channel of the first channel group are arranged as an extra sample e2n. Then, the extra words of the (2n + 1) th sample data of each channel of the first channel group are collectively arranged as an extra sample e2n + 1.

Then, the main word of the (2n) -th sample data of each channel of the second channel group is arranged together as a main sample S2n, and then the main word of each channel of the second channel group is arranged. (2
The main word of the (n + 1) -th sample data is put together and arranged as a main sample S2n + 1, and then the extra word of the (2n) -th sample data of each channel of the second channel group is put together to obtain an extra sample e2n. Then, the extra words of the (2n + 1) th sample data of each channel of the second channel group are put together to form an extra sample e.
2n + 1 (where n = 0, 1, 2,...)
).

The data structure of FIG. 16 is the same as the case 3 of FIG.
And the attribute Ar in the first channel group
In this example, the sampling frequency fs = 48 kHz and the number of quantization bits Qs = 16 bits are used as t1, and the attribute Art2 of the second channel group is fs = 48 kHz and the number of quantization bits Qs = 16 bits.

In this case, S4n and S4n + 2 are main samples of the first attribute, e4n and e4n + 2 are extra samples of the first attribute, S4n and S4n + 2 are main samples of the second attribute, e4n and e4n + 2 is the extra sample of the first attribute. Each of the first and second channel groups is basically based on two pairs of samples, and is basically based on four samples.

That is, in the data structure of FIG. 16, the signal of the first channel group which is at least two of the plurality of channels is sampled at the first frequency, and the signal of the second channel group which is another channel is sampled. The sampled data is sampled at the second frequency, and the sample data is further divided into the m1 bit main word and the LSB
M2 bit extra words.

Then, the main words of the (4n) -th sample data of each channel of the first channel group are collectively arranged as a main sample S4n, and then the (4n + 2) -th of each channel of the first channel group is arranged. Are arranged as a main sample S4n + 2, and then the extra words of the (4n) -th sample data of each channel of the first channel group are arranged as an extra sample e4n. Then, the extra words of the (4n + 2) th sample data of each channel of the first channel group are collectively arranged as an extra sample e4n + 2.

Next, the main word of the (4n) -th sample data of each channel of the second channel group is put together and arranged as a main sample S4n. 4n +
2) The main word of the second sample data is put together and arranged as a main sample S4n + 2, and then the extra word of the (4n) th sample data of each channel of the second channel group is put together into an extra sample e4n. Then, the extra words of the (4n + 2) -th sample data of each channel of the second channel group are put together to form an extra sample e4
n + 2 (where n = 0, 1, 2,...)
).

The data structure of FIG. 17 corresponds to the case 1 of FIG.
In this case, the number of quantization bits is also the first.
And the second channel group. As the attribute Art1 in the first channel group, the sampling frequency fs =
96 kHz and the number of quantization bits Qs = 20 bits are adopted, and the attribute Art2 in the second channel group is f
This is an example in which s = 48 kHz and the number of quantization bits Qs = 24 bits are used. This data structure is similar to the scalable method described above, and also has a data structure similar to the DVD video sample array structure.

That is, four samples S4n, S4n + 1, S4n + 2,
S4n + 3 is a main sample having the first attribute, and two samples S2n and S2n + 1 are main samples having the second attribute.
In this case, extra samples e4n, e4n + 1, e4n + 2, e4n + 3 exist in the first channel group, and extra samples e2n, e2n + 1 exist in the second channel group. Also in this case, the first channel group is basically based on four pairs of samples, and the corresponding second channel group is basically based on two pairs of samples, and is basically based on six pairs of samples.

By adopting the data structure as described above, the DVD audio data structure having a high-quality sound signal specification satisfying a predetermined transmission rate while retaining the type of the DVD-video audio data structure as much as possible. Obtainable.

The present invention provides a distinctive data structure.
One sampling frequency fs of the two attributes is to be a multiple of the other sampling frequency. If the two attribute groups differ only in the number of channels or the number of quantization bits, the concept of the DVD video standard can be applied to accommodate data structures with different numbers of channels and / or quantization bits. It is. For example,
This is because in the data structure shown in FIG. 4, the number of channels and / or the number of quantization bits in the attribute information of the next data following the main sample part and the extra sample part may be changed (switched) and recorded.

The present invention includes the following concept in the above data structure.

That is, FIG. 11 shows the correspondence between the time to be synchronized of each sample of the first channel group of the attribute Art1 and the second channel group of the attribute Art2, 4n, 4n + 1, 4n +.
2, 4n + 3, 4n + 4, 2n, and 2n + 1 are assigned. As can be seen from this figure, four samples are one unit. Accordingly, the four samples are handled as a single unit, and as shown in FIG. 18, two samples of the attribute Art1 and two samples of the attribute Art2 are continuously arranged.
Next, two samples of the attribute Art1 may be arranged.
This data structure corresponds to a modification of the data structure of FIG.

FIG. 19 shows a data structure according to still another embodiment. This data structure corresponds to a modification of the data structure of FIG. That is, four samples S4n, S4n + 1, S4n
+2 and S4n + 3 are main samples of the first attribute, and two samples S2n and S2n + 1 are main samples of the second attribute. In this case, extra samples e4n, e4n + 1, e4n + 2, and e4n + 3 exist in the first channel group,
Extra samples e2n, e2n + 1
Exists. Also in this case, the first channel group is basically based on four pairs of samples, and the corresponding second channel group is basically based on two pairs of samples, and is basically based on six pairs of samples.

Here, this data structure is composed of S4n, S4n + 1, e4n, e4n + of the first channel group as four pairs of samples.
1, S2n, S2n + 1, e2n, e2n + 1 of the second channel group
Is summarized. Then, two pairs of samples S4n + 2, S4n + 3, e4n + 2, and e4n + 3 of the first channel group are arranged.

When considering the sample unit described above, it can be understood as follows. That is, when the sampling frequency fs in the attribute Atr1 and the attribute Atr2 is the same (for example, FIG. 12 and FIG. 13, FIG. 15 and FIG. 16).
The number of samples after the lapse of the same time is the same in the attribute Atr1 and the channel group on the attribute Atr2. In such a case,
As in the case of the DVD video standard, data may be captured in a two-sample one-unit system.

Furthermore, the data structure of the present invention can be understood as follows. That is, the number of samples forming one unit, that is, one unit is basically 2, 4, and 6. Therefore, in order to provide versatility, data may be handled with 12 samples, which is the least common multiple of 2, 4, and 6, or 12 pairs of samples as one unit.

As described above, the number of samples per unit can be varied in various cases. In any case, the data area of the audio pack is filled for each unit. If the remaining portion of the data is less than one unit, the stuffing bytes and padding packets are filled as in the case of the video standard.

FIG. 20 shows an example in which padding packets are inserted because an area (shaded area) of less than one unit occurs. The area less than one unit refers to an area having a data amount equal to or less than a predetermined number of samples or equal to or less than a predetermined number of samples. The predetermined number of samples or the predetermined number of samples is 2, 4, 6, 12, or the like. This audio pack is 2048 bytes and is configured to always have a presentation time stamp (PTS).

In each of the above-described drawings, the attribute Art
1, the arrangement of the data of the attribute Art2 is not necessarily limited to this arrangement, and it is needless to say that the arrangement may be reversed. This arrangement may be variously changed according to the agreement.

In the above description, the sampling frequencies of 96 kHz and 48 kHz have been described. However, the present invention is not limited to this, and the sampling frequencies of 88.2 kHz and 44.1 kHz may be used. The present invention is always applicable if the relationship is double. The present invention can be easily applied if the relationship between the two frequencies is a multiple of the other with respect to the versatility.

Further, in the above description, two types of channel attributes are included in one stream, but three or more types are also applicable to the present invention.

In the above description, the data structure has been described. However, the present invention further relates to a recording medium having the above data structure, a recording method and apparatus for the recording medium, a method and an apparatus for reproducing data from the recording medium, and data transmission. It is also applicable to the method.

Next, the relationship between the overall data structure of an optical disc on which DVD audio information is recorded and the above-described audio pack will be briefly described.

FIG. 21 shows an example of the data structure of the recorded contents of the DVD audio zone (audio only title audio object set; AOTT_AOBS).

AOTT_AOBS defines a group of one or more audio objects AOTT_AOB # n. Each AOTT_AOB defines a set of one or more audio cells ATS_C # n. A program is formed by a group of one or more cells ATS_C # n, and a program chain (PGC) is formed by a group of one or more programs. This PGC is
It constitutes a logical unit for indicating the whole or a part of the audio title.

In this example, each audio cell ATS_C
# Is audio pack A_PC of 2048 byte size
It is composed of a set of K. These packs are the minimum units when performing data transfer processing. The minimum unit for performing the logical processing is a cell unit, and the logical processing is performed in this cell unit.

FIG. 22 is a view for explaining a case where a cell is accessed by the program chain information ATS_PGCI of the DVD audio zone.

According to the cell reproduction information on program # 1 in ATS_PGCI, AOB cell ATS_C #
1. ATS_C # 2 is reproduced.

If one PGC is compared to one opera,
The plurality of cells constituting the PGC can be interpreted to correspond to music or singing portions of various scenes in the opera. The contents of this PGC (or the contents of the cell)
Determined by the software provider that produces the content recorded on the disc. That is, the provider is A
Using the cell reproduction information ATS_C_PBI written in the program chain information ATS_PGCI in the TS,
Cells constituting AOTT_AOBS can be reproduced as intended.

Next, it will be described how the various rules of the first channel group and the second channel group are concretely performed on the management data.

FIG. 23 is a view for explaining the recorded contents of an audio title set (ATS) in the DVD audio zone.

The audio title set ATS includes audio title set information ATSI and an audio only title audio object set AOT.
It is composed of T_AOBS and a backup ATSI_BUP of audio title set information.

Audio title set information ATSI
Is an audio title set management table ATSI_
MAT and audio title set program chain information table ATS_PGCIT.

The audio title set program chain information table ATS_PGCIT includes an audio title set program chain information table information ATS_PGCITI and an audio title set program chain information search pointer ATS_PG.
CI_SRP and one or more audio title set program chain information ATS_PGCI.

FIG. 24 shows the recorded contents of the audio title set information management table ATSI_MAT of FIG.

That is, the audio title set information management table ATSI_MAT includes an audio title set identifier (ATSI_ID); an audio title set end address (ATS_EA); an audio title set information end address (ATSI_ATS).
_EA); version number of the adopted audio standard (VERN); end address of audio title set information management table (ATSI_MAT_EA); start address of video title set VTS for audio only title AOTT (VTS_SA); for audio only title Start address of audio object set (AOTT_AOBS_S
A) or the start address (AOTT_VOB) of the video object set for audio only title
S_SA); start address (ATS_PG) of the program chain information table for audio title set
CIT_SA); attribute of audio object set for audio only title (AOTT_AOBS)
_ATR) or attributes of the video object set for audio-only titles (AOTT_VOBS_A)
TR) # 0 to # 7; audio title set data mix coefficient (ATS_DM_COEFT) # 0 to # 1
5: Still image attribute of audio title set (ATS
_SPCT_ATR); other reserved areas are provided.

When the ATS does not have AOTT_AOBS, the start address VTS_SA of the AOTT VTS includes VTSTT_ATS used for AOTT.
The start address of the video title set VTS including VOBS is written. ATS is AOTT_AOBS
"00000000h" is this VTS_S
A is written. This is because video information may also be recorded.

The AOTT_AOBS_SA contains AT
When S has AOTT_AOBS, AO is represented by the number of logical blocks relative to the first logical block of ATS.
The start address of TT_AOBS is written. On the other hand, when the ATS does not have AOTT_ABOS, ATS
In OTT_VOBS_SA, the start address of the video object (VTSTT_VOBS) for the video title set is the VTS used for ATS.
Written with the number of logical blocks relative to the first logical block of the VTS including TT_VOBS.

The ATS_PGCIT_SA contains AT
The start address of ATS_PGCIT is written with the number of relative logical blocks from the first logical block of SI.

AOTT_AOBS_ATR which is attribute information for the audio title set or AOTT_VOB_ which is attribute information for the video title set
Eight ARTs are prepared from # 0 to # 7. AT
When S has AOTT_AOBS, the attribute of AOTT_AOB recorded in ATS is AOTT_AOBS_A.
Written to TR. On the other hand, ATS is AOTT_AOB
If it does not have S, AOTT_VOB_ART contains
VOB used for AOTT_VOB in ATS
The attributes of the audio stream in are written. This AOTT_AOBS_ATR or AOTT_VOB_
ART includes the adopted sampling frequency (44 to 1).
92 kHz) and the number of quantization bits (16 to 24 bits).

Further, in this part, a channel assignment is described.

The channel assignment describes the assignment information of each channel of the audio stream included in the video object specified by this attribute. The content of the assignment information depends on the configuration of the multi-channel. This channel assignment information
FIG. 26 to be described later. This assignment information is also described in an audio packet header described later.

The ATS_DM_COEFT is AC-
Indicates a coefficient when audio data having multi-channel output (5.1-channel output) such as 3 or DTS is mixed down to 2-channel output, and is used only for one or more AOTT_AOBs recorded in the ATS Is done. 1 if ATS does not have AOTT_AOBS
“0h” is written to all bits of each of the six (# 0 to # 15) ATS_DM_COEFTs. This one
Areas for six (# 0 to # 15) ATS_DM_COEFTs are constantly provided.

The ATS_SPCT_ATR is AOT
The attribute of the still picture stream for each still picture in T_AOBS is shown. When there is no still image in AOTT_AOBS, “0000h” is set in ATS_SPCT_ATR.
Is written. Each field of this still image is AO
This is in accordance with the information recorded in the video stream of each still image in TT_AOBS.

Each ATS_SPCT_ATR is composed of 16 bits, and 2 bits (bits b15 to b1) on the MSB side.
4) represents a video compression mode (MPEG2 or the like), and the next two bits (bits b13 to b12) are used in the TV system (N
TSC, PAL, SECAM, etc.), and the next two bits (bits b11 to b10) indicate the aspect ratio (4:
3, 16: 9, etc.) and the next two bits (bits b9 to
b8) is a display mode (4: 3 display, 16: 9 display, letter box display, etc. on a 4: 3 size TV monitor)
Is represented. The next two bits (bits b7 to b6) are reserved bits for the future. The next three bits (bits b5 to b3) represent the resolution of the still image (720 horizontal x 480 vertical in the NTSC system, 720 horizontal x 576 vertical in the PAL system). Last 3 bits on LSB side (bits b2 to b0)
Are reserved bits for the future.

FIG. 25 is a diagram for explaining the contents of the audio title set program chain information table ATS_PGCIT included in the audio title set information ATSI (the recording position of this ATS_PGCIT is written in ATS_PGCIT_SA of ATSI_MAT).

As described above, the ATS_PGCIT includes the audio title set program chain information table information ATS_PGCITI and the audio title set program chain information search pointer AT.
S_PGCI_SRP and audio title set program chain information ATS_PGCI.

The ATS_PGCI_SRP is one or more audio title set program chain information search pointers (ATS_PGCI_SRP # 1 to ATS).
_PGCI_SRP # j), and the ATS_PGC
I is the same number of ATS_PGCI_SRP and audio title set program chain information (ATS_PGC
I # 1 to ATS_PGCI # j).

Each ATS_PGCI functions as navigation data for controlling the reproduction of the audio title set program chain ATS_PGC.

[0139] Here, ATS_PGC is an audio
A unit that defines only title AOTT, AT
S_PGCI and one or more cells (A in AOTT_AOBS or A
OTT_VOBS).

Each ATS_PGCI includes general information (ATS_P) of an audio title set program chain.
GC_GI), an audio title set program information table (ATS_PGCIT), and an audio title set cell playback information table (ATS_C_
PBIT).

The ATS_PGCIT contains one or more audio title set program information (ATS_PGI).
# 1 to ATS_PGI # k), and the ATS_C_
The PBIT contains the same number of audio title set cell playback information (ATS_C_PBI # 1 to AT
S_C_PBI # k).

FIG. 26 shows channel assignment information, a first channel group and a second channel group classified based on this information.
The classification of a channel group is shown. ATSI_ in FIG.
In the MAT, attribute information of an audio object is described, and it has been described that a channel assignment exists in the attribute information. The channel assignment is the data shown in FIG.

In the case of 00000b, it means monaural, and in the case of 00001b, L, R
(Stereo) channel exists, 00
In the case of 010b, it means that Lf and Rf (left front, right front) channels exist in the first channel group, and S (surround) exists in the second channel group. In the case of 00001b, Lf,
Rf (left front, right front), and Ls, Rs (left surround, right surround) exist in the second channel group. In the case of 00100b, Lf, Rf (left front,
Light front) and LFE (low frequency effect) in the second channel group. In the case of 00101b, it means that Lf and Rf (left front and right front) exist in the first channel group, and LFE (low frequency effect) and S (surround) exist in the second channel group. In the case of 00110b, L is added to the first channel group.
f, Rf (left front, right front), LFE (low frequency effect), Ls, Rs in the second channel group
(Left surround, right surround). In the case of 00111b, the first channel group includes Lf, Rf (left front, right front),
This means that C (center) exists in the second channel group. In the case of 01000b, L is added to the first channel group.
f, Rf (left front, right front), and C (center) and S (surround) in the second channel group. In the case of 01001b, Lf and Rf (left front and right front) are assigned to the first channel group, and C (center), Ls and Rs are assigned to the second channel group.
(Left surround, right surround). In the case of 01010b, the first channel group includes Lf, Rf (left front, right front),
This means that C (center) and LFE (low frequency effect) exist in the second channel group. 01011b means that Lf and Rf (left front and right front) exist in the first channel group, C (center), LFE (low frequency effect), and S (surround) exist in the second channel group. . 01100b, the first
Lf, Rf (left front, right front) for channel group, C (center), LFE for second channel group
(Low frequency effect), Ls, Rs (left surround,
Light surround). 011
01b, Lf, Rf (left front, right front), C (center), second channel
This means that S (surround) exists in the channel group. In the case of 01110b, L is added to the first channel group.
f, Rf (left front, right front), C (center), and Ls, Rs (left surround, right surround) exist in the second channel group. 01111b, Lf,
Rf (left front, right front), C (center), and LFE (low frequency effect) exist in the second channel group. In the case of 10,000b, the first
Lf, Rf (left front, right front), C (center) in channel group, LFE in second channel group
(Low frequency effect) and S (surround) are present. In the case of 10001b, Lf, Rf (left front, right front), C
(Center), and LFE (low frequency effect), Ls, Rs (left surround, right surround) are present in the second channel group. In the case of 10010b, Lf, Rf (left front, left front,
(Right front), Ls, Rs (left surround, right surround), and LFE (low frequency effect) in the second channel group. In the case of 10011b, it means that Lf, Rf (left front, right front) and Ls, Rs (left surround, right surround) exist in the first channel group, and C (center) exists in the second channel group.

In the case of 10100b, Lf, Rf (left front, right front), Lf
s, Rs (left surround, right surround), and C (center) and LFE (low frequency effect) in the second channel group.

The attribute information shown in FIG. 24 includes AOTT_AOBS_ATR or AOTT_VOB.
_ART includes the adopted sampling frequency (44 to
192 kHz) and the number of quantization bits (16 to 24 bits).

Next, the audio pack will be described in more detail.

FIG. 27 shows the basic structure of an audio pack.

The data structure of A_PKT includes a pack header, a packet header, a substream ID, an ISRC,
The area includes a private header length, a first access unit pointer, audio data information, stuffing bytes of 0 to 7 bytes, and linear PCM audio data.

The following rules are applied to the size of the packet header. That is, if A_PKT is the first packet in the audio object, the size is 1
7 bytes, 9 bytes if the first data of the audio frame is not included, and 14 bytes otherwise.

An audio packet of the linear PCM includes a packet header, a private header, and audio data. The contents of the packet header and the private header are configured as shown in FIGS.

FIG. 28 shows a packet header. Each data is described in the order of description. Packet start code, stream id, PES packet length, "01", PES scramble control information, PES priority, data alignment indicator, copy Write, original or copy, PTS_DTS flag, ESCR_flag,
There are ES_rate flag, DSM trick mode flag, additional copy flag, PES_CRC flag, PES extension flag, PES header length. Then, 5 bytes of a description area of a presentation time stamp (PTS) indicating the reproduction time of this packet is secured. Next, a PES private data flag, a pack header field flag, a program packet order counter flag, a P_STD buffer flag, a second PES extension flag, "01", P_ST
D buffer scale and P_STD buffer size information are described.

FIG. 29 shows a private packet.

Each data is described as follows in the order of description. Substream id, reservation, ISRC number, IS
RC data, private header length, head access unit pointer, audio enhancement flag, reservation, reservation, downmix code, first quantization word length, second quantization word length, first audio sampling frequency, second audio sampling frequency , Audio reservation frequency, reservation, multi-channel type, reservation, channel assignment, dynamic range control information, and stuffing byte.

Each field item will be described as follows.

[0155] The substream id includes 10100000b indicating that it is linear PCM audio data.
Is described. ISRC used for still image control
In the number, numbers 1 to 12 indicating the range of the recorded ISRC data are described. ISRC data is I
The data specified by the SRC number is described.
As the private header length, the length is indicated by the number of logical blocks from the last byte of this field.
The first access unit pointer indicates the address of the first byte of the unit to be accessed first, using the number of logical blocks from the last byte of this field. The audio emphasis flag indicates that the emphasis is turned off when the first sampling frequency is 96 kHz or 88.2 kHz, and that the second sampling frequency is 96 kHz or 8 kHz.
The emphasis off is described also at 8.2 kHz. 0 is described for emphasis off, and 1 is described for emphasis on. In the downmix code, a coefficient table for downmixing audio samples is specified. Table number is 000
b to 1111b.

The first quantized word length describes the word length of the quantized audio samples of the first channel group. The bit length is 16 bits for 0000b, 20 bits for 0001b, and 24 bits for 0010b. Means a bit.

The second quantized word length describes the word length of the quantized audio samples of the second channel group, 16 bits for 0000b, 20 bits for 0001b, and 24 bits for 0010b. Means a bit. The case of 1111b means that it is not specified, for example, when the second channel group does not exist.

The first audio sampling frequency describes the frequency of audio samples of the first channel group. 0000b is 48 kHz, 0001b is 96 kHz, 1000b is 44.1 kHz, 1001b
Means 88.2 kHz.

The second audio sampling frequency describes the frequency of audio samples of the second channel group. 0000b is 48 kHz, 0001b is 96 kHz, 1000b is 44.1 kHz, 1001b
Means 88.2 kHz. The case of 1111b means that it is not specified, for example, when the second channel group does not exist.

[0160] In the multi-channel type, the type of the multi-channel structure of the audio sample is described. 0000b is type 1 and the others are reserved.

The channel assignment describes the state of channel assignment, and is as described with reference to FIG.

The dynamic range control information is control information for suppressing a dynamic range. The upper 3 bits of an 8-bit word indicate an integer X, and the lower 5 bits are an integer Y.
Is shown.

The linear gain is G = ^{24− (X + Y / 30)} (0 <
= X <= 7, 0 <= Y <= 29) dB, G = 24.
082-6.0206X-0.2007Y (0 <= X <
= 7,0 <= Y <= 29).

At the time of reproducing the disc, the system control unit grasps the attribute information indicating the assignment of the above-described channel group and the like, the first and second quantization word lengths of the audio data, the first and second audio sampling frequencies, and the like. By doing so, it is possible to cut out the data of the first channel group and the second channel group, and it is possible to synchronize the reproduction timing. That is, these pieces of header information can be used as synchronization information.

Next, the reproduction system of the DVD audio disc recorded as described above will be described in more detail.

[0166] Fig. 30 shows a more specific signal sequence of the reproducing apparatus for the audio stream. Data recorded on the optical disk is stored in the optical head unit 533.
And output as a high-frequency signal. The high-frequency signal (read signal) input to the system processing unit 504 is input to the synchronization detector 601. Synchronous detector 6
In step 01, a synchronization signal added to the recording data is detected, and a timing signal is generated. The read signal from which the synchronization signal has been removed by the synchronization detector 601 is input to an 8-16 demodulator 602 for demodulating 16 bits to 8 bits, and
Demodulated into a bit data string. The demodulated data is input to the error correction circuit 603, where the data is subjected to error correction processing. The error corrected data is stored in the track buffer 60.
4 to the demultiplexer 605. In the demultiplexer 605, audio packs, real-time data, and the like are identified based on the stream ID, and each pack is output to the corresponding decoder.

The audio pack is taken into the audio buffer 611. The pack header and packet header of the audio pack are stored in the control circuit 612.
Is read. The control circuit 612 recognizes the contents of the audio pack. That is, the start code, stuffing length, packet start code, stream ID, and the like of the audio pack are recognized. Further, it recognizes the packet length, the substream ID, the first access point, the audio quantization bit number, the sampling frequency, and the channel group from the channel assignment.

When such information is recognized, the control circuit 612 recognizes the packet contents of the linear PCM data and can determine the decoding method. The control circuit 612 also controls the audio buffer 6.
It is possible to know the cut-out address of the audio data for reproduction in the packet stored in the packet 11. Therefore, the audio buffer 611 is controlled by the control circuit 612, and can output the previously described samples, for example, S0, S1, e0, e1, S2, S3,. Control circuit 61
2 recognizes at least the number of quantization bits, the sampling frequency, and the channel assignment. Then, based on the recognition information, data extraction and decoder 6
13 can be set in the decode mode. This sample is supplied to a decoder 613 which performs channel processing and decodes.

FIG. 31 shows a specific configuration example of the decoder 613. A sample is supplied to an input terminal 710, and is distributed to each channel by a switch 711 based on control from a control circuit 612. L
Or, when an Lf signal (including an extra word) comes, the buffer memory 713 is sent to the buffer memory 713 when an R or Rf signal (including an extra word) comes.
14, when a signal of C (including an extra word when it comes) comes into the buffer memory 715, Ls
When the signal of Rs (including an extra word comes), the signal is distributed to the buffer memory 716. When the signal of Rs (including the extra word comes), the signal is distributed to the buffer memory 717. . Further, when an S signal arrives, the signal is distributed to the buffer memory 718. When an LEF signal arrives, the signal is distributed to the buffer memory 718.

Outputs from the buffer memories 713 to 719 are input to frame processing units 813 to 819, respectively, and are processed in frame units. Frame processing units 813 and 8
Outputs of 14, 815, 816, 817 are supplied to phase matching units 723, 724, 725, 726, 727, respectively. Also, frame processing units 815, 816, 817
Can also be supplied to the frequency converters 821, 822, 823 via the switch 820, respectively. The outputs of the frame processing units 818 and 819 are output from the frequency converter 8
24, 825.

The phase matching section is for adjusting the final phase of the signal of the first channel group and the signal of the second channel group when the second channel group is undergoing frequency conversion. Outputs of the phase matching units 723 to 727 and outputs of the frequency converters 821 to 825 are supplied to the selector 730, respectively. The selector 730 selects the signal of the corresponding channel according to the information of the channel assignment as shown in FIG.
Corresponding digital-to-analog converters 731,732,73
3, 734, 735, 736, 737.

In the above embodiment, the samples of the second channel group are frequency-converted and output. However, it is needless to say that analog conversion may be performed without performing frequency conversion. In this case, the phase matching circuit on the first channel group side may be deleted.

Next, a brief description will be given in what form the above-mentioned audio information is recorded on the optical disk.

FIGS. 32 (A), 32 (B), 32
As shown in (C) and FIG. 32 (D), when the recording surface of a part of the optical disk 10 is enlarged, a pit row is formed. This set of pits forms a sector. Therefore, a sector row is formed on the track of the optical disc. This sector is read continuously by the optical head. Then, the audio pack is reproduced in real time.

Next, one sector, for example, a sector in which audio information is described will be described. As shown in FIGS. 33A and 33B, one sector is composed of 13 × 2 frames. Then, a synchronization code is added to each frame. In the drawing, the arrangement of the frames is shown two-dimensionally, but they are recorded on the track in order from the first frame. SY0, SY5, SY1, SY5, SY5, SY5, SY5
SY2, SY5,...

The number of bits of the synchronization code and data in one frame shown in the figure is 32 bits and 1456 bits. 32 bits = 16 bits × 2, 1456 bits = 16 bits × 91. This equation means that a 16-bit modulation code is recorded. This is because, when recording is performed on the optical disc, 8-bit data is modulated into 16 bits and recorded.
Further, the sector information includes a modulated error correction code.

FIG. 34A shows one of the above physical sectors.
This shows one recording sector after 6-bit data is decoded into 8 bits. The data amount of this recording sector is
(172 + 10) bytes × (12 + 1) lines.
Each line is provided with a 10-byte error correction code. Also, there is an error correction code for one line, and this error correction code functions as an error correction code in the column direction when 12 lines are collected, as described later.

When the error correction code is removed from the data of one recording sector, a data block as shown in FIG. 34B is obtained. That is, a 6-byte sector ID, a 2-byte ID error detection code, and a 6-byte copyright management information are added to the beginning of the data to the 2048-byte main data, and a 4-byte error detection code is added to the end of the data. Data block.

The above 2048-byte data is one pack described above, and a pack header, a packet header, and audio data are described from the beginning of this one pack. Then, various guide information for processing the audio data is described in the pack header and the packet header.

As described above, one packet in which audio samples are arranged is allocated to one sector of the disk and recorded. Then, even if the information is information of one sector, the audio decoder
CM data can be favorably reproduced. This is 1
This is because the beginning of the audio data in the pack is always distributed so as to start from the beginning of the main sample. Also, this is because the pack header and the packet header describe control information sufficient for the audio decoder to process the audio data.

Next, an error correction code block (ECC block) will be described.

As shown in FIGS. 35 (A) and 35 (B), an ECC block is constituted by a set of 16 recording sectors described above. FIG. 35 (A)
Data sector of 12 rows x 127 bytes (Fig. 26 (A))
Shows a state in which 16 are collected. Then, a 16-byte outer code parity (PO) is added to each column. Each row has a 10-byte inner code parity (PI)
Is added. Further, before recording, FIG.
As shown in (B), the outer code parity (PO) of 16 bytes is distributed to each row by 1 bit. As a result,
One recording sector is configured as data of 13 (= 12 + 1) rows. In FIG. 35 (A), B0,
.., 0, B0, 1,... Indicate addresses in byte units. In FIG. 35B, 0 to 15 assigned to each block represent one recording sector. On the recording tracks of the above-mentioned disc, audio packs, management information, and optionally still image information and real-time information are arranged.

According to the present invention, the information is recorded on a disc.
Alternatively, the data structure described above is described as a data structure reproduced from a disc. However, it is easy to use the above data structure when transmitting data using a communication system. It goes without saying that a device for transmitting, a device for transmitting, and a device for receiving are also included in the category. Furthermore, in the above description, the method and the apparatus for sampling and handling the audio signal have been described.
Of course, as long as the data is used in the transmission system, it can be applied to signals other than the audio signal.

[0184]

As described above, the present invention utilizes an audio data structure standard for DVD video as much as possible and realizes an audio information recording medium having a data structure that realizes a DVD audio standard having high sound quality specifications. Equipment can be provided.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing a DVD video data sample configuration and sample arrangement relating to the present invention.

FIG. 2 is an explanatory diagram showing an example of an arrangement of DVD video packs and a configuration of an audio pack in the arrangement.

FIG. 3 is a diagram illustrating in detail a configuration of an audio pack related to a DVD video.

FIG. 4 is an explanatory diagram showing a list of examples of the data size in a packet of linear PCM data.

FIG. 5 is an explanatory diagram showing an example of generating an audio pack for a DVD video.

FIG. 6 is a diagram showing a list of linear PCM data sizes for DVD video.

FIG. 7 is a view showing a pack header of an audio pack.

FIG. 8 is a diagram showing a packet header of an audio pack.

FIG. 9 is a view for explaining a basic configuration of a disk recording / reproducing apparatus adopting scalable;

FIG. 10 is a diagram illustrating a scalable principle applied to the present invention with reference to a sample example.

FIG. 11 is a view for explaining the principle of scalability applied to the present invention by showing another sample example.

FIG. 12 is a view for explaining the principle of scalability applied to the present invention by showing still another sample example.

FIG. 13 is a view for explaining the principle of scalability applied to the present invention by showing still another sample example.

FIG. 14 is a diagram showing an example of a data sample structure according to the present invention.

FIG. 15 is a diagram showing another example of the data sample structure according to the present invention.

FIG. 16 is a diagram showing still another example of the data sample structure according to the present invention.

FIG. 17 is a diagram showing still another example of the data sample structure according to the present invention.

FIG. 18 is a diagram showing another example of the data sample structure according to the present invention.

FIG. 19 is a diagram showing another example of the data sample structure according to the present invention.

FIG. 20 is a diagram showing a simplified structure of an audio pack according to the present invention.

FIG. 21 is an explanatory diagram hierarchically showing a relationship between an audio object set and an audio pack according to the present invention.

FIG. 22 is a view shown for explaining the relationship between cells of an audio title set and program chain information according to the present invention;

FIG. 23 is an explanatory diagram showing an arrangement state of logical data of a disk on which DVD audio is recorded according to the present invention.

FIG. 24 is an explanatory diagram showing the contents of an audio title set information management table according to the present invention.

FIG. 25 is an explanatory diagram showing information forming an audio title set program chain information search pointer in FIG. 23;

FIG. 26 is a diagram shown to explain a channel assignment table according to the present invention.

FIG. 27 is a diagram showing a configuration of an audio pack according to the present invention.

FIG. 28 is an explanatory diagram showing contents of a packet header included in the audio pack of FIG. 27;

FIG. 29 is an explanatory diagram showing the contents of a private packet header included in the audio pack of FIG. 27;

FIG. 30 is a diagram showing a configuration of a disk reproducing apparatus according to the present invention.

FIG. 31 is a diagram showing an example of the internal configuration of the decoder of FIG. 30.

FIG. 32 is an explanatory diagram showing a disk, a pit row, a sector row, and a physical sector.

FIG. 33 shows the contents of a physical sector.

FIG. 34 is a diagram showing a configuration of a recording sector.

FIG. 35 is a diagram showing a configuration of an error correction code block.

[Explanation of symbols]

10: Disk, 533: Optical head unit, 502: System CPU, 504: System processing unit, 505: Data R
AM,

 ────────────────────────────────────────────────── ─── Continuing from the front page (72) Inventor Junichi Uoda 70-Yanagimachi, Saiwai-ku, Kawasaki-shi, Kanagawa F-term in Toshiba Yanagicho Plant (reference) 5C052 AA01 AB05 CC01 DD06 5C053 FA24 FA25 GB02 GB06 GB11 GB21 GB32 JA05 JA23 5D044 AB05 BC02 CC04 DE13 DE44 DE53 FG18 GK08 GK14 5D110 AA14 AA27 DA03 DA11 DA17 DC05 DE01

Claims (3)

[Claims] The data read and processed using a computer controller includes at least an audio object and management information, wherein the audio object includes a first channel group among a plurality of channels of audio signals. And a first audio data sequence (sample sequence) digitized with a first quantization bit number and a second channel or a second channel group among the audio signals of the plurality of channels. A second audio data sequence (sample sequence) digitized with a second sampling frequency and a second quantization bit number is used, and the first and second sample sequences are paired in frame units, and a frame arrangement is performed. And a frame array is housed in a pack including a header and formed as a pack row, and a set of the packs is managed as a cell, The set of cells is defined so as to be managed as a program, and the management information includes an audio title set information management table and a program chain information table necessary for reproducing the audio signals of the plurality of channels. The professional program chain information table has program chain general information, a program information table including the number of each entry cell of a plurality of programs, and a cell reproduction information table indicating a reproduction position of the cell of each program. The header includes a quantized word length of a sample of the first channel group, information of a first audio sampling frequency, and a quantized word length of a sample of the second channel group. 2 that the information of audio sampling frequency is included. Readable audio information recording medium in a computer control system to. 2. The audio information recording medium according to claim 1, wherein the header further includes dynamic range information for setting a dynamic range of reproduced audio. 3. The recording data includes at least an audio object and management information, wherein the audio object is configured to divide a first channel group among a plurality of channels of audio signals at a first sampling frequency and a first quantization. A first audio data sequence (sample sequence) digitized by the number of bits, and a second channel or a second channel group of the plurality of channels of audio signals at a second sampling frequency and a second sampling frequency; A second audio data sequence (sample sequence) digitized by the number of quantization bits is used, and the first and second sample sequences are paired in frame units, arranged in frames, and in a pack in which the frame arrangement includes a header. The packs are accommodated and formed as a row of packs. The set of packs is managed as a cell, and the set of cells is managed as a program. The management information has an audio title set information management table and a program chain information table necessary for reproducing the audio signals of the plurality of channels, and the professional program chain information table has A program information table including program chain general information, a number of each entry cell of a plurality of programs, and a cell reproduction information table indicating a reproduction position of the cell of each program, the header includes: The quantization word length of the sample of the first channel group, the information of the first audio sampling frequency, and the quantization word length of the sample of the second channel group, the information of the second audio sampling frequency are included. Data processing device for an audio information recording medium Information means for reading the data of the recording medium, and means for the from the reading obtained data to obtain the pack sequence, the first included in the header of the pack of the pack sequence
The information of the quantized word length of the sample of the channel group of the second channel group and the information of the first audio sampling frequency, and the information of the quantized word length of the sample of the second channel group and the information of the second audio sampling frequency are obtained. A control unit for obtaining a reproduction timing of the audio data in the pack in order to obtain data synchronization between the second channel groups.