ISO/IEC 14496-3:2005/Amd 9:2008

INTERNATIONAL ISO/IEC
STANDARD 14496-3
Third edition
2005-12-01
AMENDMENT 9
2008-07-01

Information technology — Coding of
audio-visual objects
Part 3:
Audio
AMENDMENT 9: Enhanced low delay AAC
Technologies de l'information — Codage des objets audiovisuels
Partie 3: Codage audio
AMENDEMENT 9: Retard faible amélioré AAC




Reference number
ISO/IEC 14496-3:2005/Amd.9:2008(E)
©
ISO/IEC 2008

---------------------- Page: 1 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.


COPYRIGHT PROTECTED DOCUMENT


©  ISO/IEC 2008
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland

ii © ISO/IEC 2008 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Foreword
ISO (the International Organization for Standardization) and IEC (the International Electrotechnical
Commission) form the specialized system for worldwide standardization. National bodies that are members of
ISO or IEC participate in the development of International Standards through technical committees
established by the respective organization to deal with particular fields of technical activity. ISO and IEC
technical committees collaborate in fields of mutual interest. Other international organizations, governmental
and non-governmental, in liaison with ISO and IEC, also take part in the work. In the field of information
technology, ISO and IEC have established a joint technical committee, ISO/IEC JTC 1.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of the joint technical committee is to prepare International Standards. Draft International
Standards adopted by the joint technical committee are circulated to national bodies for voting. Publication as
an International Standard requires approval by at least 75 % of the national bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO and IEC shall not be held responsible for identifying any or all such patent rights.
Amendment 9 to ISO/IEC 14496-3:2005 was prepared by Joint Technical Committee ISO/IEC JTC 1,
Information technology, Subcommittee SC 29, Coding of audio, picture, multimedia and hypermedia
information.

© ISO/IEC 2008 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)

Information technology — Coding of audio-visual objects
Part 3:
Audio
AMENDMENT 9: Enhanced low delay AAC
In the following, changes in existing text and tables are highlighted by grey background.

In 1.5.1.1, extend Table 1.1 with the following entries:
Table 1.1 — Audio Object Type definition based on Tools/Modules

Audio
Object
Type

…
ER AAC
39   X X X X  X X  X  X X X        X
ELD
40- (reserv-

95 ed)

Before 1.5.2, add 1.5.1.2.37:
1.5.1.2.37 Error Resilient (ER) AAC ELD object type
The enhanced low delay AAC object type (ER AAC ELD) is identical to the ER AAC LD object type, plus the
utilization of a Low Delay Filterbank (LDFB) and an enhanced low delay window. It also permits combinations
with the PNS tool as well as the Low Delay SBR tool. The ER AAC ELD object type provides the ability to
extend the usage of generic low bitrate audio coding to applications requiring a very low delay of the encoding
/ decoding chain (e.g. full-duplex real-time communications).”

In 1.5.2.4, insert the following new entries for the Baseline MPEG Surround Profile into Table 1.12
audioProfileLevelIndication and adapt the “reserved for ISO use” range accordingly (changes are highlighted
in gray):
© ISO/IEC 2008 – All rights reserved 1
Object Type ID
gain control
block switching
window shapes - standard
window shapes – AAC LD
Low Delay Window
filterbank - standard
filterbank - SSR
TNS
LTP
intensity
coupling
frequency deomain prediction
PNS
MS
SIAQ
FSS
upsampling filter tool
quantisation&coding - AAC
quantisation&coding – TwinVQ
quantisation&coding - BSAC
AAC ER Tools
ER payload syntax
EP Tool 1)
CELP
Silence Compression
HVXC
HVXC 4kbit/s VR
SA tools
SASBF
MIDI
HILN
TTSI
SBR
Layer-1
Layer-2
Layer-3
SSC (Transient, Sinusoid, Noise)
Parametric stereo
Low Delay SBR
Remark

---------------------- Page: 4 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Value Profile Level
… … …
0x33 High Efficiency AAC v2 Profile L5
0x34 Low Delay AAC Profile L1
0x35 Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L1
0x36 Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L2
0x37 Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L3
0x38 Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L4
0x39 Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L5
0x3A Baseline MPEG Surround Profile (see ISO/IEC 23003-1) L6
0x3B - 0x7F reserved for ISO use -
0x80 - 0xFD user private -
0xFE no audio profile specified -
0xFF no audio capability required -
NOTE — Usage of the value 0xFE indicates that the content described by this
InitialObjectDescriptor does not comply to any audio profile specified in ISO/IEC 14496-3.
Usage of the value 0xFF indicates that none of the audio profile capabilities are required for
this content.

In 1.6.2.1, extend Table 1.13 “AudioSpecificConfig()” as follows:
Table 1.13 — Syntax of AudioSpecificConfig()
Syntax No. of bits Mnemonic
AudioSpecificConfig ()
{
audioObjectType = GetAudioObjectType();
samplingFrequencyIndex; 4 bslbf
if ( samplingFrequencyIndex == 0xf ) {
 samplingFrequency; 24 uimsbf
}
channelConfiguration; 4 bslbf

sbrPresentFlag = -1;
psPresentFlag = -1;
if ( audioObjectType == 5 ||
 audioObjectType == 29 ) {
 extensionAudioObjectType = audioObjectType;
 sbrPresentFlag = 1;
 if ( audioObjectType == 29 ) {
  psPresentFlag = 1;
 }
 extensionSamplingFrequencyIndex; 4 uimsbf
 if ( extensionSamplingFrequencyIndex == 0xf )
  extensionSamplingFrequency; 24 uimsbf
 audioObjectType = GetAudioObjectType();
}
else {
 extensionAudioObjectType = 0;
}
switch (audioObjectType) {
2 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 5 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
case 1:
case 2:
case 3:
case 4:
case 6:
case 7:
case 17:
case 19:
case 20:
case 21:
case 22:
case 23:
 GASpecificConfig();
 break:
case 8:
 CelpSpecificConfig();
 break;
case 9:
 HvxcSpecificConfig();
 break:
case 12:
 TTSSpecificConfig();
 break;
case 13:

case 14:
case 15:
case 16:
 StructuredAudioSpecificConfig();
 break;
case 24:
 ErrorResilientCelpSpecificConfig();

 break;

case 25:

 ErrorResilientHvxcSpecificConfig();
 break;
case 26:
case 27:
 ParametricSpecificConfig();
 break;
case 28:
 SSCSpecificConfig();
 break;
case 32:
case 33:
case 34:
 MPEG_1_2_SpecificConfig();
 break;
case 35:
 DSTSpecificConfig();
 break;
case 36:
 fillBits; 5 bslbf
 ALSSpecificConfig();
 break;
case 37:
case 38:
 SLSSpecificConfig();
 break;
    case 39:
© ISO/IEC 2008 – All rights reserved 3

---------------------- Page: 6 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
       ELDSpecificConfig(channelConfiguration);

       break;
default:

 /* reserved */

}
switch (audioObjectType) {
case 17:
case 19:
case 20:
case 21:
case 22:
case 23:
case 24:
case 25:
case 26:
case 27:
   case 39:
 epConfig; 2 bslbf
 if ( epConfig == 2 || epConfig == 3 ) {

  ErrorProtectionSpecificConfig();
 }
 if ( epConfig == 3 )
  directMapping; 1 bslbf
  if ( ! directMapping ) {
  /* tbd */
  }
 }
}
if ( extensionAudioObjectType != 5 && bits_to_decode() >= 16 ) {
 syncExtensionType; 11 bslbf
 if (syncExtensionType == 0x2b7) {
  extensionAudioObjectType = GetAudioObjectType();
  if ( extensionAudioObjectType == 5 ) {
  sbrPresentFlag; 1 uimsbf
  if (sbrPresentFlag == 1) {

   extensionSamplingFrequencyIndex; 4 uimsbf
   if ( extensionSamplingFrequencyIndex == 0xf ) {
   extensionSamplingFrequency; 24 uimsbf
   }
  if ( bits_to_decode() >= 12 ) {
   syncExtensionType; 11 bslbf
   if (syncExtensionType == 0x548) {
   psPresentFlag; 1 uimsbf
   }
  }
  }
 }
}
}





4 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
In 1.6.2.2.1, extend Table 1.15 “Audio Object Types” as follows:
Table 1.15 — Audio Object Types
Object Audio Object definition of elementary stream Mapping of audio payloads to access
Type ID Type payloads and detailed syntax units and elementary streams
… … … …
39 ER AAC ELD ISO/IEC 14496-3 subpart 4 see subclause 1.6.2.2.2.4

Add 1.6.2.2.4 with the title "ER AAC ELD":
1.6.2.2.2 .4 ER AAC ELD
The top level payload for ER AAC ELD is defined in er_raw_data_block_eld(). All definitions mentioned in
subclause 1.6.2.2.2.3 are also valid for this AOT.

At the end of 1.6.5.1, add a sentence:
“NOTE: None of these signaling methods described in this subclause is allowed for AAC ELD in order to
signal the low delay sbr tool. For this case the ldSbrPresentFlag in the ELDSpecificConfig is to be used”

Before 4.5, insert Tables AMD9.2 to AMD9.8:
Table AMD9.2 — Syntax of top level payload for audio object types ER AAC ELD
(er_raw_data_block_eld())
Syntax No. of bits Mnemonic
er_raw_data_block_eld(channelConfiguration)
{
switch(channelConfiguration) {
 case 1:
  single_channel_element_eld();
  break;
 case 2:
  channel_pair_element_eld ();
  break;
 case 3:
  single_channel_element_eld ();
  channel_pair_element_eld ();
  break;
 case 4:
  single_channel_element_eld ();
  channel_pair_element_eld ();
  single_channel_element_eld ();
  break;
© ISO/IEC 2008 – All rights reserved 5

---------------------- Page: 8 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
 case 5:
  single_channel_element_eld ();
  channel_pair_element_eld ();
  channel_pair_element_eld ();
  break;
 case 6:
  single_channel_element_eld ();
  channel_pair_element_eld ();
  channel_pair_element_eld ();
  lfe_channel_element_eld ();
  break;
 case 7:
  single_channel_element_eld ();
  channel_pair_element_eld ();
  channel_pair_element_eld ();
  channel_pair_element_eld ();
  lfe_channel_element_eld ();
  break;
 default:
  /* reserved */
  break;
}
if (ldSbrPresentFlag) {
 er_low_delay_sbr_block(channelConfiguration);
}
cnt = bits_to_decode() / 8;
while ( cnt >= 1 ) {
 cnt -= extension_payload(cnt);
}
byte_alignment();
}


Table AMD9.3 — Syntax of single_channel_element_eld()
Syntax No. of bits Mnemonic
single_channel_element_eld()
{
individual_channel_stream_eld (0);
}

Table AMD9.4 — Syntax of lfe_channel_element_eld()
Syntax No. of bits Mnemonic
lfe_channel_element_eld()
{
individual_channel_stream_eld (0);
}

6 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Table AMD9.5 — Syntax of channel_pair_element_eld()
Syntax No. of bits Mnemonic
channel_pair_element_eld()
{
common_window = 1;
 max_sfb;     6 uimsbf
ms_mask_present; 2 uimsbf
if ( ms_mask_present == 1 ) {
 for (sfb = 0; sfb < max_sfb; sfb++) {
  ms_used[0][sfb]; 1 uimsbf
 }
}

individual_channel_stream_eld (common_window);
individual_channel_stream_eld (common_window);

}

Table AMD9.6 — Syntax of individual_channel_stream_eld()
Syntax No. of bits Mnemonic
individual_channel_stream_eld (common_window)
{
global_gain; 8 uimsbf
if (! common_window) {
 max_sfb;     6 uimsbf
}
section_data ();
scale_factor_data ();
tns_data_present; 1 uimsbf
if (tns_data_present) {
 tns_data ();
}
if (! aacSpectralDataResilienceFlag) {
 spectral_data ();
}
else {
 length_of_reordered_spectral_data; 14 uimsbf
 length_of_longest_codeword; 6 uimsbf
 reordered_spectral_data ();
}
}

© ISO/IEC 2008 – All rights reserved 7

---------------------- Page: 10 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Table AMD9.7 — Syntax of er_low_delay_sbr_block
Syntax No. of bits Mnemonic
er_low_delay_sbr_block(channelConfiguration)
{
switch (channelConfiguration) {
  case 1:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 2:
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 3:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 4:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 5:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 6:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  break;
 case 7:
  low_delay_sbr_data(ID_SCE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  low_delay_sbr_data(ID_CPE, ldSbrCrcFlag, bs_amp_res);
  break;
 default:
  /* reserved */
  break;
}
}

8 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 11 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Table AMD9.8 — Syntax of low_delay_sbr_data()
Syntax No. of bits Mnemonic
low_delay_sbr_data(id_aac, ldSbrCrcFlag, bs_amp_res)
{
 if (ldSbrCrcFlag) {
  bs_sbr_crc_bits; 10 uimsbf
 }

 if (bs_header_flag) { 1
  sbr_header();
 }

 If (id_aac==ID_SCE) {
  sbr_single_channel_element(bs_amp_res);
 } else if (id_aac==ID_CPE) {
  sbr_channel_pair_element(bs_amp_res);
 }
}

Note 1: bs_amp_res is in general transmitted inside sbr_header() function enclosed in the
ELDSpecificConfig(). But the parameter can be updated by the sbr_header() function here.

Extend Table 4.142 as followings:
Table 4.142 — AAC error sensitivity category assignment for extended payload and low delay sbr
payload
6 - drc_band_top dynamic_range_info()
6 - drc_bands_incr dynamic_range_info()
6 - drc_bands_present dynamic_range_info()
6 - drc_bands_reserved_bits dynamic_range_info()
6 - drc_tag_reserved_bits dynamic_range_info()
6 - dyn_rng_ct dynamic_range_info()
6 - dyn_rng_sgn dynamic_range_info()
6 - excluded_chns_present dynamic_range_info()
6 - pce_instance_tag dynamic_range_info()
6 - pce_tag_present dynamic_range_info()
6 - prog_ref_level dynamic_range_info()
6 - prog_ref_level_present dynamic_range_info()
6 - prog_ref_level_reserved_bits dynamic_range_info()
6 - additional_excluded_chns excluded_channels()
6 - exclude_mask excluded_channels()
5 - extension_type Extension_payload()
5 - data_element_version Extension_payload()
7 - fill_byte Extension_payload()
7 - fill_nibble Extension_payload()
7 - other_bits Extension_payload()
8 - dataElementLengthPart Extension_payload()
8 - data_element_byte extension_payload()
© ISO/IEC 2008 – All rights reserved 9
extension_payload
low delay sbr payload
data_element
Function

---------------------- Page: 12 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
9 - bs_sbr_crc_bits sbr_extension_data()
9 - bs_header_flag sbr_extension_data()
9 - bs_fill_bits sbr_extension_data()
9 9 bs_amp_res sbr_header()
9 9 bs_start_freq sbr_header()
9 9 bs_stop_freq sbr_header()
9 9 bs_xover_band sbr_header()
9 9 bs_reserved sbr_header()
9 9 bs_header_extra_1 sbr_header()
9 9 bs_header_extra_2 sbr_header()
9 9 bs_freq_scale sbr_header()
9 9 bs_alter_scale sbr_header()
9 9 bs_noise_bands sbr_header()
9 9 bs_limiter_bands sbr_header()
9 9 bs_limiter_gains sbr_header()
9 9 bs_interpol_freq sbr_header()
9 9 bs_smoothing_mode sbr_header()
9 9 bs_data_extra sbr_single_channel_element()
9 9 bs_reserved sbr_single_channel_element()
9 10 bs_add_harmonic_flag sbr_single_channel_element()
9 10 bs_extended_data sbr_single_channel_element()
9 10 bs_extension_size sbr_single_channel_element()
9 10 bs_esc_count sbr_single_channel_element()
9 10 bs_extension_id sbr_single_channel_element()
9 9 bs_data_extra sbr_channel_pair_element()
9 9 bs_reserved sbr_channel_pair_element()
9 9 bs_coupling sbr_channel_pair_element()
9 10 bs_add_harmonic_flag sbr_channel_pair_element()
9 10 bs_extended_data sbr_channel_pair_element()
9 10 bs_extension_size sbr_channel_pair_element()
9 10 bs_esc_count sbr_channel_pair_element()
9 10 bs_extension_id sbr_channel_pair_element()
9 - bs_frame_class sbr_grid()
9 - tmp sbr_grid()
9 - bs_freq_res sbr_grid()
9 - bs_pointer sbr_grid()
9 - bs_var_bord_0 sbr_grid()
9 - bs_var_bord_1 sbr_grid()
9 - bs_num_rel_0 sbr_grid()
9 - bs_num_rel_1 sbr_grid()
9 9 bs_df_env sbr_dtdf()
9 9 bs_df_noise sbr_dtdf()
9 10 bs_invf_mode sbr_invf()
9 10 bs_env_start_value_balance sbr_envelope()
9 10 bs_env_start_value_level sbr_envelope()
9 10 bs_codeword sbr_envelope()
9 10 bs_noise_start_value_balanc sbr_noise()
e
9 10 bs_noise_start_value_level sbr_noise()
9 10 bs_codeword sbr_noise()
9 10 bs_add_harmonic sbr_sinusoidal_coding()
- 9 bs_frame_class sbr_ld_grid()
- 9 tmp sbr_ld_grid()
- 9 bs_freq_res sbr_ld_grid()
- 9 bs_transient_position sbr_ld_grid()
- 9 bs_sbr_crc_bits low_delay_sbr_data()
- 9 bs_header_flag low_delay_sbr_data()
10 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 13 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
At the end of 4.5.5, add the following new Figure:

SCE CPE CPE LFE SBR SBR SBR
ext. payloads
(SCE) (CPE) (CPE)
Anc Data bit
L R L R
stuffing
4a 4b 4c 4d 4e 4f
3f
3a 3b 3c 3d 3e
2a 2b 2c 2d 2e 2f 10a 10b 10c 8a 7a
1a 1b 1c 1d 1e 1f 9a 9b 9c 5a 5b
0a 0b

Figure AMD9.1 — Dependency structure in case of ER multichannel AAC ELD syntax
(channelConfigurati == 6)

After 4.6.18, add a new subclause with the title "Low Delay SBR":
4.6.19 Low Delay SBR
4.6.19.1 Introduction
Low Delay SBR is derived from the standard SBR tool in order to be utilized as a bandwidth extension coder
in communication scenarios. Thus, the algorithmic delay of this tool is minimized to achieve an overall delay
low enough for bi-directional communication applications.
Summary of modifications:
• Frame length adopted to a core codec with 512 or 480 samples per frame
• Frame-locked time/frequency grid
• Minimization of delay in QMF buffer
• Utilizing a Complex Low Delay Filterbank

The low delay SBR tool is defined by the following modifications with respect to the standard algorithm (i.e.
SBR audio object type). All clauses/subclauses can be found in the brackets after each headline.

4.6.19.2 Definitions, Constants and Variables
4.6.19.2.1 Definitions (changes to 4.6.18.2.1.20)
• time slot: finest resolution in time for SBR envelopes and noise floors. One time slot equals one
subsample in the QMF domain.
© ISO/IEC 2008 – All rights reserved 11

---------------------- Page: 14 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
4.6.19.2.2 Constants (changes to 4.6.18.2.5)
• RATE: For core codec AAC ELD the constant RATE= 1 should be used instead of RATE= 2 .
4.6.19.2.3 Variables (changes to 4.6.18.2.6)
• numTimeSlots: The number of time slots for core codec AAC ELD is numTimeSlots= 16 , for 512
AAC frame and numTimeSlots= 15 , for 480 AAC frame.
• t : Due to the removed additional delay the offset of the HF-generation module is set to
HFGen
t = 2. The following Figure Figure AMD9.2 explains the offsets for core codec AAC ELD
HFGen
correctly.
numTimeSlots⋅RATE−+12
numTimeSlots⋅RATE−+12

Figure AMD9.2 — Synchronization for low delay SBR
4.6.19.2.4 Inverse Filtering (changes to 4.6.18.6.2)
Due to the modified buffer management the calculation of the covariance matrix has to be changed as follows:
The upper limit of n is to be changed from numTimeSlots⋅RATE+ 6−1 to numTimeSlots⋅RATE−1.
4.6.19.3 Time-Locked frequency grid
The time/frequency grid for the Low Delay SBR is specified in the following subclauses.
4.6.19.3.1 FrameClasses
Low Delay SBR uses a reduced set of frame classes which are listed in the Table below:
Table AMD9.9 — bs_frame_class
bs_frame_class Meaning
0 FIXFIX
1 LD_TRAN

4.6.19.3.2 sbr_ld_grid()
Low Delay SBR uses a different syntax to signal the grid data. The standard sbr_grid() syntax is replaced by
sbr_ld_grid(), as defined in the following:
12 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 15 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Table AMD9.10 — Syntax of sbr_ld_grid
Syntax No. of bits Mnemonic
sbr_ld_grid(ch)
{
switch (bs_frame_class) { 1 uimsbf
 case FIXFIX:
  bs_num_env[ch] = 2^ tmp; 2 uimsbf,
Note 1
  if (bs_num_env[ch] == 1)
           bs_amp_res; 1 uimsbf
  bs_freq_res[ch][0]; 1
  for (env = 1; env < bs_num_env[ch]; env++)
  bs_freq_res[ch][env] = bs_freq_res[ch][0];

  break;

 case LD_TRAN:

  bs_transient_position 4 uimsbf
  bs_num_env[ch] =  Note 3
  LD_Envelope_Table[bs_transient_position][num_envelopes];
  for (env = 0; env < bs_num_env[ch]; env++)
  bs_freq_res[ch][env]; 1
  break;

if (bs_num_env[ch] > 1)
 bs_num_noise[ch] = 2;

Else
 bs_num_noise[ch] = 1;


}

Note 1: bs_num_env is restricted according to subclause 4.6.18.3.
Note 2: the Table LD_Envelope_Table is given in Table AMD9.11

4.6.19.3.3 Calculation of t (l) (changes to 4.6.18.3.3)
E
in case bs_frame_class = LD_TRAN:
t (0)= 0
E
t (L )=numberTimeSlots
E E
t (l)= LD _EnvelopeTable[bs _transient _ position][l+1]for0 E E
4.6.19.3.4 Calculation of l (changes to 4.6.18.7.2)
A
in case bs_frame_class = LD_TRAN:
l = LD _EnvelopeTable[bs _transient _ position][4]
A
© ISO/IEC 2008 – All rights reserved 13

---------------------- Page: 16 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
4.6.19.3.5 Envelope Lookup Table
Table AMD9.11 — Lookup Table for LD_Envelope_Table (bs_transient_position)
bs_transient_position num_envelopes border[1] border[2] transientIdx
0 2 4 - 0
1 2 5 - 0
2 3 2 6 1
3 3 3 7 1
4 3 4 8 1
5 3 5 9 1
6 3 6 10 1
7 3 7 11 1
8 3 8 12 1
9 3 9 13 1
10* 3/2 10 14/- 1
11 2 11 - 1
12 2 12 - 1
13 2 13 - 1
14 2 14 - 1
15 2 15 - 1
* in case of AAC frame = 480 use second Table entry

4.6.19.4 Low Delay SBR Filterbank (changes to 4.6.18.4)
Instead of subclause 4.6.18.4, use the description below for filterbank processing. Note, that basically the
processing descibed in the flowcharts Figure 4.41, 4.42, 4.43, 4.50, 4.51, 4.52 is similar to the processing of
the low-delay filterbank. Only the windowing and the modulation differ.

4.6.19.4.1 Analysis filterbank
• Shift the samples in the array x by 32 positions. The oldest 32 samples are discarded and 32 new
samples are stored in positions 0 to 31.
• Multiply the samples of array x by the coefficient of window ci. The window coefficients ci are obtained by
linear interpolation of the coefficients c, i.e. through the equation
1
ci(i)=[]c(2i+1)+c(2i) , 0≤i< 320
2
The window coefficients of c can be found in Table AMD9.17.
• Sum the samples according to the formula in the flowchart in Figure 4.41 to create the 64-element array
u.
14 © ISO/IEC 2008 – All rights reserved

---------------------- Page: 17 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
Calculate 32 new subband samples by the matrix operation Mu, where
0≤ k< 32
i⋅π⋅ (k+ 0.5)⋅ (2⋅n− 95) ⎧
⎛ ⎞
Μ(k,n)= 2⋅ exp ,
⎜ ⎟ ⎨
64 0≤ n< 64
⎝ ⎠
⎩
In the equation, exp() denotes the complex exponential function and i is the imaginary unit.

4.6.19.4.2 Synthesis filterbank
• Shift the samples in the array v by 128 positions. The oldest 128 samples are discarded.
The 64 new complex-valued subband samples are multiplied by the matrix N, where
0≤ k< 64
1 i⋅π⋅ (k+ 0.5)⋅ (2⋅n− 63) ⎧
⎛ ⎞
Ν(k,n)= ⋅ exp ,
⎜ ⎟
⎨
64 128 0≤ n< 128
⎝ ⎠
⎩
In the equation, exp() denotes the complex exponential function and i is the imaginary unit. The real part
of the output from this operation is stored in the positions 0 to 127 of array v.
• Extract samples from v according to the flowchart in Figure 4.42 to create the 640-element array g.
• Multiply the samples of array g by window c to produce array w. The window coefficients of c can be
found in Table AMD9.17.
• Calculate 64 new output samples by summation of samples from array w according to the last step in the
flowchart of in Figure 4.42

4.6.19.4.3 Downsampled synthesis filterbank
• Shift the samples in the array v by 64 positions. The oldest 64 samples are discarded.
The 32 new complex-valued subband samples are multiplied by the matrix N, where
0≤ k< 32
1 i⋅π⋅ (k+ 0.5)⋅ (2⋅n− 31) ⎧
⎛ ⎞
Ν(k,n)= ⋅ exp ,
⎜ ⎟
⎨
64 64 0≤ n< 64
⎝ ⎠
⎩
In the equation, exp() denotes the complex exponential function and i is the imaginary unit. The real part
of the output from this operation is stored in the positions 0 to 63 of array v.
• Extract samples from v according to the flowchart in Figure 4.43 to create the 320-element array g.
• Multiply the samples of array g by the coefficient of window ci to produce array w. The window
coefficients ci are obtained by linear interpolation of the coefficients c, i.e. through the equation
1
ci(i)=[]c(2i+1)+c(2i) , 0≤i< 320
2
The window coefficients of c can be found in Table AMD9.17.
• Calculate 32 new output samples by summation of samples from array w according to the last step in the
flowchart of Figure 4.43



© ISO/IEC 2008 – All rights reserved 15

---------------------- Page: 18 ----------------------
ISO/IEC 14496-3:2005/Amd.9:2008(E)
4.6.19.5 Low Power SBR Filterbank (4.6.18.8.2)
4.6.19.5.1 Real-valued analysis filterbank
• Shift the samples in the array x by 32 positions. The oldest 32 samples are discarded and 32 new
samples are stored in positions 0 to 31.
• Multiply the samples of array x by the coefficient of window ci. The window coefficients ci are obtained by
linear interpolation of the coefficients c, i.e. through the equation
1
ci(i)=[]c(2i+1)+c(2i) , 0≤i< 320
2
The window coefficients of c can be found in Table Table AMD9.17.
• Sum the samples according to the formula in the flowchart to create the 64-element array u.
• Calculate new 32 subband samples by the matrix operation M u, where
r
0≤ k< 32
⎧
⎛π⋅ (k+ 0.5)⋅ (2⋅n− 95)⎞
Μ (k,n)= 2⋅ cos ,
⎜ ⎟
⎨
r
64 0≤ n< 64
⎝ ⎠
⎩
4.6.19.5.2 Real-valued synthesis filterbank
• Shift the samples in the array v by 128 positions. The oldest 128 samples are discarded.
The 64 new subband samples are multiplied by the matrix N , where
r
⎧0≤ k< 64
1 π⋅ (k+ 0.5)⋅ (2⋅n− 63)
⎛ ⎞
Ν (k,n)= ⋅ cos ,
⎜ ⎟
⎨
r
32 128 0≤ n< 128
⎝ ⎠
⎩
The output from this operation is stored in the positions 0 to 127 of array v.
• Extract samples from v according to the flowchart in Figure 4.50 to create the 640-element array g.
• Multiply the samples of array g by window c to produce array w. The window coefficients of c can be
found in Table Table AMD9.17.
• Calculate 64 new output samples by summation of samples from array w according to the formula in the
flowchart of Figure 4.50.
4.6.19.5.3 Downsampled real-valued synthesis filterbank
• Shift the samples in the array v by 64 positions. The oldest 64 samples are discarded.
The 32 new subband samples are multiplied by the matrix N , where
r

0≤ k< 32
⎧
1 ⎛π⋅ (k+ 0.5)⋅ (2⋅n− 31)⎞
Ν (k,n)= ⋅ cos ,
⎜ ⎟
⎨
r
32 64 0≤ n< 64
⎝ ⎠
⎩

The output from this operation is stored in the positions 0 to 61 of array v.
• Extract samples from
...