MXPA99006859A - A method and device for interfacing variable-rate sampled digital audio information to a string of uniform-sized blocks, and a unitary medium so produced by a write-interfacing - Google Patents

Abstract

For mapping sampled digital audio information onto a linear physical mapping space that is partitioned in a string of uniform-sized blocks, in particular for variable-rate coded audio information that is distributed over successive audio units which are each composed from one or more audio packets, each block is supplemented with a block header for indicating an actual content of the block in question with respect to any constituent packet of the audio information.

Description

A METHOD AND A DEVICE FOR INTERCONNECTING INFORMATION FROM DIGITAL AUDIO SAMPLED AT A VARIABLE SPEED TO A ORDERLY SEQUENCE OF UNIFORM SIZE BLOCKS, AND A UNIT MEDIUM SO PRODUCED THROUGH AN INTERCONNECTION WRITTEN The invention relates to a method, as set forth in the preamble of Claim 1. The current state of the art of digital audio coding is the well-known Compact Disc system, with its ramifications. The progress in storage technology and audio coding technology allows increasing the amount of audio information on a unit medium according to the dimensions of the DC (Disks).

Compact) standard. A particular characteristic is the variable speed coding, which, however, requires an easily accessible indicator organization to have access to the subsequent reading of the ordered sequence or chain of Audio Units. Accordingly, among other things, the object of the present invention allows a reading device to correctly recover all the information belonging to an audio unit that could have been scattered over several blocks or storage sectors. Now, therefore, In accordance with one of its aspects, the invention is characterized, as set forth in the characterizing part of Claim 1. The invention also relates to a unit storage means produced by practicing such a method, and with a reading device for reading a unit storage medium thus produced. Additional advantageous aspects of the invention are set forth in the appended Claims. These and the additional aspects and advantages of the invention will be discussed in greater detail hereinafter, with reference to the description of the preferred embodiments, and in particular with reference to the Attached figures that show: Figures 1, 1, a record carrier, Figure 2, a reproducing device, Figure 3, a recording device, Figure 4, a diagram of a linear physical storage space; Table 1, a storage format according to the invention; Table 2, a syntax of an audio stream; Table 3, a header format; Table 4, a list of data type specification; Table 5, an audio block header syntax; * Table 6, a packet information syntax; Table 7, another definition of the type of data. The Figure shows a record carrier in the form of disk 11 with disk 19 and a central hole 10. Track 19 is arranged in a spiral pattern of turns that form substantially parallel tracks on an information layer. The carrier can be an optical disk with a layer of recordable or precoded information. Examples of a recordable disc are the CD-R, CR-RW and DVD-RAM, while an audio CD is a pre-registered disc. Pre-registered discs can be manufactured by first registering a master disc and then pressing "discs for the consumer. The track 19 on the registration carrier t-: recordable is indicated by providing a pre-stamped track structure during the manufacture of the blank record carrier. The track can be configured as a pre-slot 14, which allows a read / write head Track 19 while scanning. The information is recorded on the information layer by means of optically detectable marks along the track, for example, *** »depressions and outgoing. Figure Ib is a section along line b-b of the recordable record carrier 11, wherein the transparent substrate 15 contains the recording layer 16 and the protective layer 17. The pre-race 14 can be applied as an indentation, an elevation, or as an appropriate material that deviates from its surroundings . For convenience of use, audio information about the record carrier has been subdivided into units of information, which usually last for a few minutes, for example, songs on an album or movements of a symphony. Usually, the record carrier also contains access information to identify the units of information, such as the so-called table of contents (TOC), or included in a file system such as ISO 9660 for CD-ROM. . The access information may include the execution time and address of each unit of information, and also additional information, such as the title of the song. The audio information is recorded in digital representation after analog-to-digital (A / D) conversion. Examples of A / D conversion in the PCM of 16 bits per sample at 44.1 kHz, known for audio CDs and Sig to Delta modulation of 1 bit at a higher sampling rate, for example, so-called bitstream 64 x Fs. The latter method represents a high-quality encoding method, with the choice between high-quality decoding and low-quality decoding, the latter allowing for a simpler decoding circuit. Reference is made in this regard to the publications "A digital decimator filter for the conversion of analog to digital high-fidelity audio signals", by "JJ van der Kam, document D5 infra, and" A higher-order topology for interpolator modules for oversampling A / D converters ", by Kírk CH Chao et al, document D6. After the A / D conversion, the digital audio is compressed to audio data at a speed of bits variable for the record on the information layer. The compressed audio data is read from the record carrier at such a rate that, after decompression, substantially the original time scale will be restored. • ^ when the audio information is played continuously.

Accordingly, the compressed data must be retrieved from the record carrier at a rate that depends on the v * H 1 variable bit rate. The data is retrieved from the record carrier at the so-called transfer rate, i.e. the data byte transfer rate of the record carrier to a decompressor. The record carrier can have a uniform spatial data density, which gives the highest data storage capacity per unit area. In such a system, the transfer speed is proportional to the relative linear velocity between the middle and the read / write head. If a buffer is provided before the decompressor, the actual transfer rate is the speed before that buffer. Figure 2 shows a reproducing apparatus according to the invention for reading a record carrier 11 of the type shown in Figure 1. The device has motor means 21 for rotating the record carrier 11 and a read head 22 for scanning the track of the record carrier. The positioning means 25 effect the approximate radial position of the reading head 22. The reading head comprises a known optical system with a radiation source for generating a beam 24 which is guided through optical elements and focused to a point 23. on a track of the information layer. The read head further comprises a focusing actuator for moving the focus of the radiation 24 along the optical axis of the beam and a tracking actuator for fine positioning of the point 23 in a radial direction on the center of the track . The tracking actuator may comprise coils for moving an optical element or may be arranged to change the angle of a reflective element. The radiation reflected by the information layer is detected by a known detector in the read head 22, for example, a four-quadrant diode, to generate a read signal and additional detector signals, including tracking error and error signals. Focus error for the tracking and focusing actuators, respectively. The read signal is processed by the reading means 27 to recover the data, and reading means, which are of a usual type, for example, comprising a channel decoder and an error corrector. The recovered data is passed to data selection means 28, to select the compressed audio data to be passed over a buffer 29. The selection is based on indicators of the type of data also recorded on the record carrier, for example headers in a framed format. From the buffer 29, the compressed audio data is passed to the decompressor 31 as the signal 30. This signal can also be sent to an external decompressor. The decompressor 31 decodes the compressed audio data to reproduce the original audio information at the output 32. The decompressor can be connected separately, for example in a high-quality autonomous digital-to-analog audio converter (D / A converter), as indicated by the dotted rectangle 33 in Figure 2. Alternatively, the buffer can be placed before the data selection means. The buffer 29 can be placed in a separate housing or can be combined with a buffer in the decompressor. The device further has a control unit 20 for receiving control commands from a user or from a central computer not shown, which, via control lines 26, such as a collective conductor of the system, are connected to the motor means 21, the positioning means 25, the reading means 27 and the data selection means 28, and possibly also to a collective conductor 29 for the control of the level saturation of the collective driver. For this purpose, the control unit 20 may comprise a control circuit, such as a microprocessor, a program memory and control gates, to perform the procedures described below. The control unit 20 can be implemented as a state machine for logic circuits. The compression and decompression of audio of a suitable type are known. The audio can be compressed after being digitized, analyzing the correlation in the signal, and producing parameters for fragments of a specific size. During the decompression process is used contrary to the reconstruction of the original signal. If the original digitized signal is exactly reconstructed, it is said that the (de) compression has no losses, while the (de) compression with losses will not reproduce certain details of the original signal, however, they are not substantially detectable by the ear or human eye. Most known audio and video systems, such as DCC or MPEG, use lossy compression, while lossless compression is used to store computer data. Examples of audio compression and decompression can be found in D2, D3 and D4, hereinafter, of which in particular the lossless compression of D2 is suitable for high-quality audio. The data selection means 28 are arranged to download any packaged data, which has been added during registration. When the control unit 20 is instructed to play an audio information unit of the record carrier, the positioning means 25 is controlled by placing the read head on the portion of the track containing the TOC. The start address for that information unit will then be recovered from the TOC via the data selection means 28. Alternatively, the content of the TOC can be read only once and stored in a memory when the disk is inserted into the device. . To reproduce the information unit, the motor means 21 are controlled to rotate the record carrier at the appropriate speed. The required rotation speed can be denoted as such, to adjust the motive means. Alternatively, the rotation speed can be adjusted through time codes that are stored together with the audio data to indicate the associated playing duration. To provide continuous playback without a subflow or overflow of the buffer, the transfer rate is coupled to the reproduction rate of the D / A converter, i.e., to the bit rate after decompression. For this purpose, the apparatus may comprise a reference frequency source for controlling the decompressor and the rotation speed may be set dependent on the reference frequency and the velocity profile. Alternatively or additionally, the rotation speed can be adjusted using the average fill level of the buffer 29, for example, by slowing down the rotation speed when the buffer is filled more than 50% on average. Figure 3 shows a recording device for writing information according to the invention on a record carrier 11 of a type, which is (re) writable. During a write operation, marks are formed representing the information about the record carrier. The marks can be in any optically readable form, for example, in the form of areas whose reflection coefficient differs from their surroundings, through registration in materials such as dyes, alloys or phase changes, or in the form of areas with a magnetization direction different from that of its surroundings, when the magneto-optical material is recorded. The writing and reading of information for the registration of optical discs and the useful rules for formatting, error correction and channel coding are well known, for example, from the CD system. The marks can be formed through a spot or spot 23 generated on the recording layer, via a beam 24 of electromagnetic radiation, usually a laser diode. The recording device comprises similar basic elements as defined with reference to Figure 2, that is, a control unit 20, motor means 21 and positioning means 25, but has a distinctive writing head 39. The audio information is presents on the inlet of the compression means 35, which can be placed in a separate housing. Proper compression has been described in D2, D3 and D4. The compressed audio at a variable bit rate on the output of the compression means 35 is passed to the buffer 36. From the buffer 36, the data is passed to data combining means 37 to aggregate packaged data and additional control data. The total data stream is passed to the writing means 38 for registration. The writing head 39 is coupled to the writing means 38, which comprise, for example, a formatting device, an error encoder and a channel encoder. The data presented at the input of the writing means 38 is distributed to the logical and physical sectors, according to the rules for formatting and coding, and converted to writing signals for the writing head 39. The unit 20 is arranged to control the buffer 36, the data combining means 37 and the writing means 38, via the control line 26 and to perform the positioning procedure as described above for the reading apparatus. Alternatively, the recording apparatus can be arranged to read, having the characteristics of the reproductive apparatus and a combined writing / reading head. Figure 4 is a diagram of a linear physical storage space. The upper trace 50 shows the distribution of the audio flow in the so-called Audio Units. For Audio Units, analog audio can be sampled to produce 1-bit to 44.1-kHz signals, which is a standard sampling frequency for a Compact Disc. When the multiplicity is 64 times, the stereo quality requires a data rate of approximately 5.6 Mbits / second. A more compact format is achieved through audio coding with lower losses, to increase the storage density by a factor of about 2, but as shown in Figure 1, this will produce Nl to N + 2 Audio Units which may have mutually non-uniform sizes. On the other hand, the storage on a unitary medium, such as a disk or tape, the transmission via a physical transmission means, is preferably organized in compartments having mutually uniform dimensions, which has been indicated by means of blocks or sectors Ml. to M + 4 on line 54. To allow quick access to the blocks, each block has a header h, which during the reading will obviate the need to grammatically analyze the audio flow. Several blocks, such as the M and M + l blocks, accommodate an audio package of only a single Audio Unit, as in this case, the Audio Unit N. However, a maximum storage density is achieved as shown through the union of several audio packets in a single storage block, such as the union of audio packets Nl, 1 and N, 0 in block M, and also packets N, 3, N + l, 0 and N + 2, 0 in block M + 3. In the Figure, the packages as shown in line 52, have as their first index the number of their Audio Unit, and as a second index, the number within their audio block (advancing from 0 upwards). As shown, the packages have a maximum size, so that one package will always fit a single block of standard size. On the other hand, the minimum size of a package falls to a granularly applicable level of storage per blog. The number of packages per Audio Unit has a higher limit that is determined solely by the maximum size of an Audio Unit. Table 1 shows the storage format according to the invention, for the same configuration as shown in Figure 4. Here, each column belongs to a single block M to M + 4. Each block begins with a header area, which may have a non-uniform size. In addition, each block contains a whole number of packets that may have mutually non-uniform sizes. In addition to the Audio Units, the store may contain one or more Supplemental Data Units as an accessory for a particular Audio Unit, and one or more fill or offset units, as an additional accessory for a particular Audio Unit. The filling or compensation makes the bit rate constant, and represents simulated data to maintain a proper degree of buffering. The supplementary data may belong to a layer of arbitrary functionality, such as the subcode. A Unit or an Audio frame can start at any position within a particular block. The Audio Units can transgress the edge of a Block, and in the modality, an Audio Unit in general will be larger than a Block. However, an Audio Unit can be so short that it could fit into a single Block. A single Block could, therefore, contain the starting point of a plurality of Audio Units. A frame is related to the actual duration of the audio during playback, namely 1/75 of a second. After the audio, it contains different information that belongs to your audio. A sector also has a whole number of packages. Table 2 illustrates the syntax of an audio stream according to the invention, written in the well known Computer Language C. The first part is related to the Audio Multiplex Flow that contains a specification of the Loop of Audio Blocks numbered 0..N. Note that in this description, N indicates an arbitrary parameter. The number of bits (column on the right) of the block in question is defined by the length of the block. The second part of the Table is again in C, and it is related to a single Audio Block that contains an Audio Block Header and a loop specification of Packs numbered 0 ... N. The number of bits of the packet in question is defined by its length. As discussed, the data may have one of a plurality of respective data types. Table 3 shows a header format of a preferred but exemplary embodiment according to the invention, again written in C. The bit numbers of the different parts have been specified in the column on the right. The first bit indicates if the block in question contains the starting point of an Audio Unit. If it is positive, the following 48 bits specify several parameters of this Audio Unit, namely: a single bit that indicates the effective start of this Audio Unit, a time code of 30 bits to be used by a reading device for perform functions such as skipping a pre-specified amount of time. The second part of the header is always present. In the first part, specify the distance measured in the number of blocks, until 15, until the next Audio Unit begins, to allow a logical jump to the next Audio Unit. Each unit is linked to a single time code, and vice versa. Functionally, storage can be multiplexed between audio units, packed units, and supplementary data units. Accordingly, going to a particular unit can be done simply by waiting for a predetermined time interval until the storage area of this next unit passes, through the continuous motor movement of a storage medium, such as a disk. Frequently, skipping a track in a transversal way will accelerate this process, but even if this is applied, for some reasons such cross-track skipping would not be allowed. In addition, the header specifies the number of packets within the block by 3 bits. Next, for each such packet, there is a loop specification of the data type of that packet through 5 bits, and of that length in 11 bits. Also the number of packets is determined, therefore, by the code for the length of the header in question. In general, there is a two-level management organization: first, the correct block sector is directed through the following indicator of the unit. Subsequently, the local addressing is carried out, through the header of the local block indicating the address, such as through the signaling of the lengths of one more packets. Table 4 shows the specification of the data type through the 5 bits indicated here for Table 3. Several of the 32 combinations have been reserved. Five are used to specify various types of coding. One indicates the occurrence of the next CD. The rest have been reserved. Table 5 shows the syntax of an audio block header. The names of the different units of information, the numbers of bits, the format, and if applicable, the values have been indicated. The frame information may contain a time code. Note that the indicator of the next unit in Table 3 has been deleted. Table 6 shows the information syntax of the package. The names of the different units of information, the lengths in bits, the format, and if applicable, the values have been indicated. Table 7 shows a definition of the type of different data, as a variation of Table 4. The different types of audio packets can now be defined in the applicable_TOC area. Note that CD TEXT corresponds to a supplementary data package.

List of related documents (DI) Reserch Disclosure number 36411 August 1994, pages 412-413 (D2) PCT / IB97 / 01156 (PHN 16.452) ADC 1 bit and lossy audio compression (D3) PCT / IB97 / 01303 (PHN 16.405) Audio compression (D4) EP-A 402.973 (PHN 13.241) Audio compression (D5) "A digital decimator filter for analog-to-digital conversion of high-fidelity audio signals", by JJ van der Kam at Philips Techn. Rev. 42, no. 6/7, April 1986, pp. 230-8. (D6) _ "A higher-order topology for modular interpolators for oversampling A / D converters", by Kirk C.H. Chao et al in IEEE Trans. in Circuits and Systems, Vol 37, no. 3, March 1990, pp. 309-18.

Block Block M + l Block M + 2 Block M + 2 Block M + 4 Table 1 #bits audio_mux_fluj or (). { for (? = 0; i <N_blocks_audio; n ++ ¡ { audio_block () length audio_block } } #bits audio_block (). { header_block_audio () for (i = 0;? <N_packets; i ++). { package (data type [i]) package_ length [i]} } Table 2 #bits header_block_audio (). { no start audio unit if (sin_inicio_unidad_audio = 0). { package_country_unit 1 code_time 30} next_unit 4 N_packs 3 for (i = 0; i <N_packets; i ++). { datatype [i] 5 length [i] 11} Table 3 Type Data content 00000-00111 Reserved 01000 Stereo Fs DSD 64 NOT encoded No Losses 01001-10101 Reserved - 10110 Stereo Fs DSD 48 No Losses Coded 10111 Multiple Channel Fs DSD 48 No Loss Coded 11000 Stereo Fs DSD 64 No Loss Coded 11001 Multiple Channel Fs DSD 64 No Loss Coded 11010-11011 Reserved 11100 CD TEXT 11101-11110 Reserved 11111 Packing Table 4 tbits format value Header_Audio (). { N-Packages 3 Umt3 1..7 N_Start_Block 3 Uint3 1..7 Reserved 1 Uintl 0 Coded_Sin Losses 1 Untl for (p = l; p < = N_Packs; p ++) Info Pack () [p] 16 Info_ Pack for (f = l; f < = N_Banks_Inits; f ++) Info_Blocks () [f] 24 or 4 í Info_ Package Table 5 #bits value format Info_Package [p]. { Start_Block 1 Uintl 1. .0 Reserved 1 Uintl 0 Type_Data [p] 3 Type_ 2, 3, 7 Data Length_Package [p] 11 Uintl 1. .2045 Table 6 Table 7

Claims (5)

1. CHAPTER CLAIMEDICATORÍO
2. Having described the invention, it is considered as a novelty and, therefore, the content is claimed in the following
3. CLAIMS: 1. A method for mapping the digital audio information, sampled on a linear physical topographic space that is distributed in an ordered sequence or chain of blocks of uniform size, characterized in that the audio information encoded at variable speed, is distributed over successive audio units, each comprising one or more audio packets, including in each of the blocks a block header to indicate the current content of the block in question, with respect to any constituent packet containing audio information or contain other information. 2. The method according to claim 1, characterized in that it allows and tries to pack units and / or supplementary data units in the same way as the audio units. 3. The method according to claim 1, characterized in that the topographic space is a linear physical storage space.
4. 4. The method according to claim 1, characterized in that each block header points to the start of the next audio unit.
5. 5. The method according to claim 1, characterized by the header of the block uniformly indicates to each Data Packet a type of Data and a Packet and Length. H.H. The method in accordance with the claim 1, characterized in that the header indicates if its block contains a starting point of a particular Audio Unit and if it is positive, it specifies a time code in the Audio Unit in question. 7. The device, characterized in that it is arranged to execute a method according to claim 1. 8. A unit storage means, characterized in that it is produced to execute a method according to claim 1, or by a device in accordance with 7. The means according to claim 8, characterized in that it is executed by an optical disk. 10. A device arranged to read a storage medium according to claim 8, characterized in that the device comprises reading means for reading a sequence of blocks, detection means for detecting multiple headers that collectively indicate several packages that constitute a single unit. of audio, and mounting means for mounting the audio unit in question of its packages for presentation to a user. A device according to claim 10, characterized in that it comprises retaining means for retaining and actuating the means, and access means for accessing the medium and connecting the reading means.