1、 INTERNATIONAL STANDARD ISO/IEC 23003-2:2010 TECHNICAL CORRIGENDUM 2 Published 2014-09-01 INTERNATIONAL ORGANIZATION FOR STANDARDIZATION ORGANISATION INTERNATIONALE DE NORMALISATION INTERNATIONAL ELECTROTECHNICAL COMMISSION COMMISSION LECTROTECHNIQUE INTERNATIONALEInformation technology MPEG audio t
2、echnologies Part 2: Spatial Audio Object Coding (SAOC) TECHNICAL CORRIGENDUM 2 Technologies de linformation Technologies audio MPEG Partie 2: Codage dobjet audio spatial (SAOC) RECTIFICATIF TECHNIQUE 2 Technical Corrigendum 2 to ISO/IEC 23003-2:2010 was prepared by Joint Technical Committee ISO/IEC
3、JTC 1, Information technology, Subcommittee SC 29, Coding of audio, picture, multimedia and hypermedia information. ICS 35.040 Ref. No. ISO/IEC 23003-2:2010/Cor.2:2014(E) ISO/IEC 2014 All rights reserved Published in Switzerland ISO/IEC 23003-2:2010/Cor.2:2014(E) 2 ISO/IEC 2014 All rights reserved2R
4、eplace the following text in the subclause “7.7.2.4 Stereo to binaural “x-2-b“ processing mode” (inappropriate indexing): * 1212 . v DDEDD with: * 0101 v 2 DDE DD Replace the following text in the subclause “7.7.2.4 Stereo to binaural “x-2-b“ processing mode” (inappropriate and confusing indexing):
5、,1212 x x j x i ij ij iijj d d ee dddd with: , , , 0, 1, 0, 1, xj xi x ij ij iijj d d ee dddd Replace the following text in the subclause “7.7.2.4 Stereo to binaural “x-2-b“ processing mode” (redundant definition of already introduced variables and inappropriate indexing): The downmix matrix of size
6、 1 with elements can be found as , lx D N , lx i d0.1 0.05 ,1 0.1 10 10 11 0 l i l i l i DCLD DMG l i DCLD d , 0.05 ,2 0.1 1 10 11 0 l i l i DMG l i DCLD d . The stereo downmix matrix of size 2 with elements l D N , l x i d can be found as , , ll x xii dd . with: where the downmix matrix x D of size
7、 1 and of size N D 2 N contain elements , x i d , . 0,1 x Replace the following text in the subclause “7.6.2.4 Derivation of the MPS parameters” (incorrect statement: “parameters are quantized, formatted as MPS bitstream”): The obtained CLD, ICC and ADG parameters are quantized, formatted as MPS bit
8、stream and fed into the MPS decoder. with: ISO/IEC 23003-2:2010/Cor.2:2014(E) ISO/IEC 2014 All rights reserved 3The obtained , and parameters are fed into the MPS decoder. ADG D CLD D ICC DRemove the following equation and its lead-in text in the subclause “7.6.3.3 Stereo Preprocessing” (this equati
9、on contradicts to the one given above, namely ): 2 1,2 00 , 00 ,. r diag otherwise P vw 2 Rd and finally the mix matrix, 1 1 2 0 0 w Pvv w d 2RR 2 dReplace the following text in the subclause “7.6.3.3 Stereo Preprocessing” (incomplete specification and inconsistent notation ): 1 P G Mod and the mix
10、matrix G is given as: Mod 1,2 ,0 , ,. diag r otherwise gG G G vec Mod Similarly, the mix matrix is given as: P 2 2 1,2 00 , 00 ,. r diag otherwise P vw 2 Rd with: The SAOC downmix preprocessor unit modifies the downmix according to the MPS parameters derived from SAOC data and rendering information.
11、 The matrices G , and of size 1 P P 2 22 are given as 2 1,2 , ,. diag r otherwise gG G G vec11 00 PG 1 , ISO/IEC 23003-2:2010/Cor.2:2014(E) 4 ISO/IEC 2014 All rights reserved 2 1,2 00 , 00 ,. r diag otherwise P vw 2 RdReplace the term: “section” with: “subclause” in the text. Replace the following t
12、ext in the subclause “7.13.4 Low Power SAOC processing” (incorrect statement “transcoding (and stereo output) case”; incorrect equation: “ ”): 1,2 0 r in the stereo downmix transcoding (and stereo output) case, always forcing in section 7.6.3.3. 1,2 0 r with: in the stereo downmix case, always forci
13、ng 2 1,2 r in subclause 7.5.5.1.1. Remove the following text in the subclause “7.7.2 Downmix processor” (redundant text; incorrect statement for the SAOC transcoding mode “yielding the final output PCM signal”): For the decoder mode of the SAOC system, the output signal of the downmix processor (rep
14、resented in the hybrid QMF domain) is fed into the corresponding synthesis filterbank as described in ISO/IEC 23003-1:2007 yielding the final output PCM signal. The downmix processing incorporates the mono, stereo and, if required, subsequent binaural processing. Note: The following text exists for
15、the SAOC decoding mode: The SAOC parameter processing unit calculates the upmix parameters ( G , and ) derived from the SAOC data (OLD, IOC, DMG, DCLD), rendering (and HRTF) information. The downmix processor applies these parameters and uses the corresponding synthesis filterbank yielding the final
16、 output PCM signal. 1 P P 2Replace the following text in the subclause “7.6.2.2 SAOC processing” (redundant and missing description): Associated with each OTT element is the channel level difference (CLD) parameter that describes the relative level differences between the two output channels, and th
17、e inter-channel correlation (ICC) parameter for the desired cross-correlation between the output signals. ISO/IEC 23003-2:2010/Cor.2:2014(E) ISO/IEC 2014 All rights reserved 5In contrast to the MPS the SAOC bitstream comprises the relative level of each audio object in the downmix signal, termed the
18、 object level difference (OLD) and the inter-object correlation (IOC). The task of the SAOC parameter processing unit is to estimate all CLD and ICC parameters from the SAOC data and the rendering matrix. This process is performed for each OTT element independently. with: The Channel Level Differenc
19、e (CLD) and Inter Channel Correlation (ICC) parameters associated with each OTT element describe the relative level differences and cross-correlation between two output channels, correspondently. The Channel Prediction Coefficient (CPC) parameters associated with TTT box represent Prediction Coeffic
20、ients needed for reconstruction of the third output signal from two input ones. The Arbitrary Downmix Gains (ADG) parameters describe the modification of the downmix signal. Replace the following text in the subclause “7.6.3.2.1 Stereo downmix (“x-2-5”) processing mode” (redundant text; missing equa
21、tion: term J is undefended; incorrect definition: “ 1 JD E D”; confusing notation): To avoid numerical problems when calculating the term 1 JD E D, Jis modified. First the eigenvalues 1,2 of J are calculated, solving 1,2 det( ) 0 JI. Eigenvalues are sorted in descending ( 12 ) order and the eigenvec
22、tor corresponding to the larger eigenvalue is calculated according to the equation above. It is assured to lie in the positive x-plane (first element has to be positive). The second eigenvector is obtained from the first by a 90 degrees rotation: . * 1 12 12 2 0 0 Jv v v v with: where 1 JD E D . The
23、 matrix of size is defined as J 22 , * 1 12 12 2 0 0 J JJ JJ J Jvv vv where both modified eigenvalues J i are determined as max , J JJ ii c , 1 80 J c . The eigenvalues 1,2 J ( 12 J J ) of the matrix J and corresponding eigenvectors 1,2 J v are calculated solving the following characteristic equatio
24、ns 1,2 det 0 J DED I , ISO/IEC 23003-2:2010/Cor.2:2014(E) 6 ISO/IEC 2014 All rights reserved 1,2 1,2 0 JJ DED I v . Replace the following equation: 1,1 1,2 * 2,1 2,2 3 3,1 3,2 tt tt tt TA D I . with: 1,1 1,2 1 2,1 2,2 3 3,1 3,2 tt tt tt TA D . Replace the following equation: 1 1 min , 2 r w d1 d , 2
25、 min , 2 r w d2 d2 , with: 12 22 1,1 2,2 min , 2 , min ,2 max , max , RR dd rr w d . Replace the following equation: 2 1,1 11 0 2 5,5 max , 10log max , f CLD f , 2 1,5 1,5 1 22 1,1 5,5 max , max , max , ff ICC ff , with: 2 1,1 11 0 2 5,5 max , 10log max , f CLD f , 2 1,5 1 22 1,1 5,5 max , max , max
26、 , f ICC ff , Replace the following text in “6.1 Payloads for SAOC”: for (i=0; ibsNumDmxChannels + 1; i+) idxPDGi = EcDataSaoc(PDG, i, numBands); Note 1 with: ISO/IEC 23003-2:2010/Cor.2:2014(E) ISO/IEC 2014 All rights reserved 7for (i=0; ibsNumDmxChannels + 1; i+) idxPDGi = EcDataSaoc(DCLD, i, numBa
27、nds); Note 1 Replace the following text in “6.1 Payloads for SAOC”: ote 1: XXX is to be replaced by the value of dataType (OLD, IOC, NRG, DCLD, DMG, PDG). Nwith: Note 1: XXX is to be replaced by the value of dataType (OLD, IOC, NRG, DCLD, DMG). Replace the following text in “7.1.2 Dequantization of
28、the SAOC parameters”: , 01 pi N 01 psL , 01 proc pb M . Table 49 Dimensions and value ranges of the parameter indices idxXXX Parameter idxOLD idxNRG idxIOC idxDMG idxDCLD idxPDG Dimension pspi pipspb pspb pipipspb pspi pspi Value range 0 . 15 0 63 0 . 7 -15 . 15 -15 . 15 -15 . 15 with: , 01 pi N 01
29、psL , 01 proc pb M , 01 pd M . Table 49 Dimensions and value ranges of the parameter indices idxXXX Parameter idxOLD idxNRG idxIOC idxDMG idxDCLD idxPDG Dimension pspbpd pipspb pspb pipipspb pspi pspi Value range 0 . 15 0 63 0 . 7 -15 . 15 -15 . 15 -15 . 15 eplace the following text in “7.1.2 Dequan
30、tization of the SAOC parameters”: case DCLD, DMG, PDG: with: Rswitch (XXX) if (bsQuantCoarseXXXpisetIdx) ISO/IEC 23003-2:2010/Cor.2:2014(E) 8 ISO/IEC 2014 All rights reservedswitch (XXX) case DCLD, DMG: oarseXXXpisetIdx) Repl parameters”: switch (XXX) oarseXXXpisetIdx) with:case DCLD, DMG: oarseXXXp
31、isetIdx) Repl eters”: idxXXXpipspb = 0; if (bsQuantC ace the following text in “7.1.2 Dequantization of the SAOCcase DCLD, DMG, PDG: if (bsQuantCswitch (XXX) if (bsQuantC ace the following text in “7.1.2 Dequantization of the SAOC paramswitch (XXX) case NRG, DCLD, DMG, PDG: with:switch (XXX) case NRG, DCLD, DMG: idxXXXpipspb = 0;
