hash stringlengths 32 32 | doc_id stringlengths 7 13 | section stringlengths 3 121 | content stringlengths 0 2.2M |
|---|---|---|---|
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.5.2 Data dependencies | The AC-4 bitstream syntax supports data to be encoded incrementally over multiple consecutive codec frames (dependency over time). The bitstream syntax conforming to the present document supports individual dependency over time for substreams. If the transmitted data of one codec frame has no dependency over time for a... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.6 Decoder compatibilities | Decoders conforming to the present document shall be capable of decoding bitstreams where 0 ≤ bitstream_version ≤ 2. Decoders conforming to the present document shall be capable of decoding all presentations that conform on the one hand to the decoder compatibility level and, on the other hand, conform to either ETSI T... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.7 Decoding modes | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.7.1 Introduction | The present document specifies two decoding modes - full decoding and core decoding. The decoder implementation shall support at least one of these two modes. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.7.2 Full decoding mode | The full decoding mode supports the full immersiveness in audio experience and the highest resolution of spatial information. Using this decoding mode the decoding tools Advanced Coupling and Advanced Joint Channel Coding decode all coded channels and enable the immersive channel configurations to be played back with t... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.7.3 Core decoding mode | The core decoding mode enables the immersive experience with a reduced number of channels and the object-audio experience with reduced spatial information to support decoding on low-complexity platforms. Using this decoding mode, decoding tools such as Advanced Coupling, Advanced Joint Channel Coding or Advanced Joint ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8 Decoding process | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.1 Overview | This clause describes how the decoder, specified in this documentation, shall decode the AC-4 bitstream by utilizing the AC-4 decoding tools specified in clause 5. This process implies the usage of the AC-4 bitstream elements that are specified and described in clause 6. The general process that shall be performed to d... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.2 Selecting a presentation | The selection of presentations is application-dependent. To successfully select an appropriate presentation and the related substreams, the decoder may execute the following steps. 1) Create a list of presentations available in the bitstream. a) Initially derive the number of presentations, n_presentations, from the bi... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3 Decoding of substreams | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.1 Introduction | In a presentation where presentation_version = 1, two types of substreams containing audio content can be present: • Channel Audio Substream (CAS) (b_channel_coded = 1) • Object Audio Substream (OAS) (b_channel_coded = 0) Presentations with presentation_version=0 support CAS coding only. The decoding of each substream ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.2 Identification of substream type | Substreams can be assigned to the categories: Main, Music & Effects, Associate or Dialogue. The Main and the Music & Effects substreams require the same processing, hence they are denoted as MME substreams from here onwards. Substream types can be identified using the following information: • If presentation_version = ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.3 Substream decoding overview | Figure 4: Decoding of a Substream ETSI ETSI TS 103 190-2 V1.1.1 (2015-09) 33 The conditions shown in Figure 4 are specified in Table 6. Table 6: Conditions in flow chart for substream decoding No. Condition Description 1.1 if (spec_frontend == 0) Tracks with an associated spectral frontend type of ASF (spec_frontend=0)... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.4 Decoding of object properties | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.4.1 Introduction | Decoding of object properties present in the OAMD portions of the bitstream shall comprise the following steps: 1) Retrieve and decode the object audio metadata from the different locations in the bitstream. - Determine the number and type of objects present in the currently decoded substream group by decoding the OAMD... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.4.2 Object Audio Metadata location | Object properties are coded in multiple OAMD portions inside the bitstream, which are present on different bitstream layers. This saves bandwidth, because some of the transmitted properties are common to all objects, some are common to a group of objects, and some describe properties for individual objects. The locatio... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.5 Spectral Frontend(s) | The Spectral Frontend (SF) is the first tool that shall be utilized to decode a present substream. The SF decodes the data present in sf_data and provides a block of spectral lines for an audio track and associated information about the subsequent windowing process for the frequency-to-time transformation. sf_data elem... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.6 Stereo and Multichannel Processing (SMP) | The Stereo and Multichannel tool (SMP) shall be utilized to process the output tracks of the Audio Spectral Frontend (ASF). If the input channel configuration is one of the immersive channel configurations, the decoder shall utilize the SMP tool for immersive input as specified in clause 5.2. For all other input channe... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.7 Inverse Modified Discrete Cosine Transformation (IMDCT) | After SSF processing or ASF processing plus SMP, the decoder shall perform a frequency-to-time transformation utilizing the IMDCT tool specified in ETSI TS 103 190-1 [1], clause 5.5. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.8 Simple Coupling (S-CPL) | If the decoder processes an immersive_channel_element where the immersive_codec_mode is either SCPL or ASPX_SCPL, the decoder shall utilize the S-CPL tool specified in clause 5.3. For the processing of all other channel elements or object audio the S-CPL tool shall be bypassed. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.9 QMF Analysis | The output of the IMDCT tool (subsequently S-CPL processed if applicable) shall be transformed into the QMF spectral domain by the QMF analysis tool described in ETSI TS 103 190-1 [1], clause 5.7.3. The output of the QMF analysis for each track is one matrix Q(ts, sb), where 0 ≤ ts < num_qmf_timeslots and 0 ≤ sb < num_... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.10 Companding | The companding tool is specified in ETSI TS 103 190-1 [1], clause 5.7.5. When decoding a channel audio substream containing a single_channel_element, channel_pair_element, 3_0_channel_element, 5_X_channel_element or 7_X_channel_element, the decoder shall utilize the companding tool as specified in ETSI TS 103 190-1 [1]... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.11 A-SPX | For coding configurations specified in this clause, an AC-4 decoder shall process the QMF domain signals except for the LFE channel signal with the A-SPX tool specified in ETSI TS 103 190-1 [1], clause 5.7.6. ETSI TS 103 190-1 [1], clause 6.2.10 specifies how the A-SPX tool shall be utilized to process single_channel_e... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.12 Advanced Joint Channel Coding (A-JCC) | For the full decoding mode of a channel audio substream containing immersive_channel_element, where immersive_codec_mode is ASPX_AJCC, the decoder shall utilize the A-JCC tool for full decoding mode as specified in clause 5.6.3.5.2. For the core decoding mode of a channel audio substream containing immersive_channel_el... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.13 Advanced Joint Object Coding (A-JOC) | For the full decoding mode of an object audio substream containing A-JOC coded content (b_ajoc is 1), the decoder shall utilize the A-JOC tool as specified in clause 5.7. The input to the A-JOC tool is the output of the preceding A-SPX tool. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.14 Advanced coupling - A-CPL | If the AC-4 decoder operates in full decoding mode, the decoder shall utilize the Advanced Coupling tool. For decoding of immersive_channel_element, the A-CPL tool is specified in clause 5.5.2. Which channels the decoder shall process is shown in Table 13. For decoding of 22_2_channel_element no A-CPL processing is req... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.15 Dialogue Enhancement | The operation for the application of Dialogue Enhancement is performed by different processes depending on the decoding mode and the substream type - CAS or OAS. For CAS processing the decoder shall utilize the DE processing tool depending on the input channel configuration and the coding mode as specified in Table 14.... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.16 Direct DRC bitstream gain application | When decoding a channel audio substream with coded DRC gain values present in the bitstream (drc_compression_curve_flag = 0), the decoder shall decode the gain values for the channel configurations listed in ETSI TS 103 190-1 [1], clauses 4.3.13.7.1 and 5.7.9.3.2 respectively. The decoder shall utilize the channel grou... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.17 Substream gain application for operation with associated audio | This clause describes the application of gains related to associated audio for presentations that include an associated audio substream. Location of the associated audio gains If the selected presentation has a presentation_version = 1 and b_associated is set to 1 in the ac4_presentation_substream associated with this ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.18 Substream gain application for operation with dialogue substreams | This clause describes the application of gains related to dialogue for presentations that include a dialogue substream. Location of the dialogue gains If the selected presentation has a presentation_version = 1 and b_dialog is set to 1 in the ac4_presentation_substream associated with this presentation, the gains relat... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.3.19 Substream Rendering | Substream rendering describes the process of rendering the decoded channels or object essences to the output channel configuration. This process is significantly different for the two types of available substreams - CAS and OAS. Channel Audio Substream (CAS) The decoder shall utilize the Channel Renderer tool specified... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.4 Mixing of decoded substreams | This clause describes the process of applying substream group gains to the channel configuration instances of the selected presentation as well as mixing these into a single output channel configuration instance. The channel configuration instances are the output of the substreams renderers, as described in clause 4.8.... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5 Loudness correction | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5.1 Introduction | The loudness of the audio signal is determined by the dialnorm value. The AC-4 bitstream syntax supports presence of additional loudness correction data for the cases: • downmixing to a lower channel configuration • decoding of an alternative presentation • realtime loudness correction data (derived from realtime loudn... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5.2 Dialnorm location | For presentations with presentation_version = 1, the dialnorm value shall be extracted from the ac4_presentation_substream. For presentations with presentation_version = 0, the dialnorm value shall be derived: • from the basic_metadata of the associated substream for presentations containing associated audio, and • fro... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5.3 Downmix loudness correction | When downmixing is done in the decoder, the loudness shall be adjusted using the output channel-specific loudness correction factor from the loud_corr element that relates to the selected downmix. EXAMPLE: When transmitting 7.1.4 content and downmixing it to 7.1, the loudness correction factor loud_corr_gain_7_X shall ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5.4 Alternative presentation loudness correction | When decoding an alternative presentation, i.e. an AC-4 presentation with b_alternative = 1, a target-specific loudness correction shall be applied: target_corr,chts,sb = 2 target_corr_gain × ,chts,sb where 0 ≤ ts < num_qmf_timeslots and 0 ≤ sb < num_qmf_subbands Here, sin,ch and starget_corr,ch refer to th... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.5.5 Realtime loudness correction data | When realtime loudness correction data rtll_comp_gain is present in the bitstream, this loudness correction factor shall be applied as: rtll_comp,chts,sb = 10 rtll_comp_gain × in,chts,sb where 0 ≤ ts < num_qmf_timeslots and 0 ≤ sb < num_qmf_subbands Here, sin,ch and srtll_comp,ch refer to the input and output... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.6 Dynamic Range Control | NOTE: Only gains originating from compression curves are applied in this clause. Directly transmitted gains are applied in the substream decoding process as specified in clause 4.8.3.16. The decoder shall utilize the Dynamic Range Control (DRC) tool specified in ETSI TS 103 190-1 [1], clause 5.7.9 and DRC metadata drc_... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.7 QMF Synthesis | A QMF synthesis filter shall transform each audio channel from the QMF domain back to the time domain as specified in ETSI TS 103 190-1 [1], clause 5.7.4. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 4.8.8 Sample rate conversion | For values of frame_rate_index not equal to 13, the following requirements apply • The decoder shall be operated at external sampling frequencies of 48 kHz, 96 kHz, or 192 kHz. • The decoder shall utilize the frame rate control tool specified in clause 5.11 to adjust the sampling frequency to the external sampling freq... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5 Algorithmic details | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.1 Bitstream processing | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.1.1 Introduction | Bitstream tools assemble, multiplex, or select the correct parts of the bitstream for processing. ETSI ETSI TS 103 190-2 V1.1.1 (2015-09) 43 All the substream tools refer to bits that are parsed from the bitstream. Before substream parsing can commence, preparatory steps sometimes need to be taken: • Clause 5.1.2 speci... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.1.2 Elementary Stream Muxing Tool | The Elementary Stream Muxing (ESM) tool combines substreams from multiple bitstreams according to presentation structure and selection. AC-4 allows distributing a presentation over multiple elementary streams. The ESM tool takes multiple AC-4 bitstreams as input, and selects the appropriate substreams of each to presen... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.1.3 Efficient High Frame Rate mode | AC-4 is capable of aligning the frame rates with video signals of up to 120 fps. To improve encoding efficiency at high frame rates, the present document introduces an efficient high frame rate mode. The efficient high frame rate mode is enabled on a per-presentation basis. In this mode, the codec receives raw_ac4_fram... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.1 Introduction | This tool extends the SMP tool as specified in ETSI TS 103 190-1 [1], clause 5.3 by introducing additional processing modes for the channel_data_elements specified in clause 6.2.4. For all the channel_data_elements specified in ETSI TS 103 190-1 [1], clause 4.2.6, SMP shall be applied according to ETSI TS 103 190-1 [1]... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.2 Interface | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.2.1 Inputs | The input to the stereo and multichannel processing tool are scaled spectral lines of tracks derived from decoding the sf_data element stored in: • a channel_pair_element (nSAP = 2); • a 3_0_channel_element (nSAP = 3); • a 5_X_channel_element (nSAP = 5); • a 7_X_channel_element (nSAP = 7); • an immersive_channel_elemen... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.2.2 Outputs | The outputs from the extended stereo and multichannel processing tool are nSAP blocks of scaled spectral lines: sSMP,[A''|B''|C''|…] nSAP vectors of blk_len spectral lines assigned to intermediate decoding signals. In most cases these signals are implicitly assigned to channels (L, R, C, ...) with discrete speaker loca... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.2.3 Controls | The bitstream and additional information used by the stereo audio processing tool is: blk_len block length. Equal to the number of input and output spectral lines in one channel. sap_used[g][sfb] array indicating the operating mode of the stereo audio processing tool for group g and scale factor band sfb. sap_gain[g][s... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.3 Processing the immersive_channel_element | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.3.1 Introduction | The immersive_channel_element enables the immersive channel configurations listed in clause A.3, in Table A.31 through Table A.41. The immersive_channel_element provides a number of audio tracks derived from different combinations of channel data elements (see ETSI TS 103 190-1 [1], clause 5.3.3), similar to the 5_X_ch... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.3.3 immersive_codec_mode = ASPX_ACPL_2 | This clause defines processing when immersive_codec_mode = ASPX_ACPL_2. 1) Tracks O0, O1, ..., O6 shall be produced as specified in ETSI TS 103 190-1 [1], clause 5.3.3. 2) Tracks A, B, C, D, E, F, G shall be assigned to tracks A' through G' as specified in step 2 in clause 5.2.3.2. 3) The rest of the tracks shall be fi... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.3.4 immersive_codec_mode = ASPX_AJCC | This clause defines processing when immersive_codec_mode = ASPX_AJCC. 1) Tracks O0,O1,...,O4 shall be produced as specified in ETSI TS 103 190-1 [1], clause 5.3.3. 2) Tracks A,B,C,D,E shall be assigned to tracks A' through E' as specified in step 2 in clause 5.2.3.2, where b_5fronts = 0. ETSI ETSI TS 103 190-2 V1.1.1 (... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.2.4 Processing the 22_2_channel_element | The 22_2_channel_element enables the channel configuration for 24 channels (including two LFE channels) as described in Table A.42. The 22_2_channel_element comprises two mono_data elements and 11 two_channel_data elements. These channel data elements shall be processed according to ETSI TS 103 190-1 [1], clause 5.3.3.... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3 Simple Coupling (S-CPL) | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.1 Introduction | The Simple Coupling tool operates in the time domain, processing the output of the IMDCT. The Simple Coupling tool is used on signals that are coded in an immersive_channel_element when immersive_codec_mode ∈ {SCPL, ASPX_SCPL}. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.2 Interface | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.2.1 Inputs | XinSCPL,[A|B|…] nSCPL,in time domain signals, each being an IMDCT processed output signal of the SMP tool. The XinSCPL signals each consist of frame_length PCM audio samples. The number of S-CPL input signals, nSCPL,in, depends on b_5fronts as indicated in Table 23. Table 23: Number of S-CPL input signals b_5fronts nSC... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.2.2 Outputs | XoutSCPL,[a|b|…] nSCPL,out decoupled time signals. The XoutSCPL signals each consist of frame_length PCM audio samples. The number of S-CPL output signals, nSCPL,out, equals nSCPL,in. For core decoding, nSCPL,out is limited to a maximum of 7 channels. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.3 Reconstruction of the output channels | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.3.1 Full decoding | In full decoding mode, the output channels (L, R, C, ...) of the Simple Coupling tool are created using a multiplication of a matrix MSCPL with the eleven or thirteen IMDCT processed output signals (A'', B'', C'', ...) of the Stereo and Multichannel tool (see Table 21). The descriptors (A'', B'', C'', ...) are re-used ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.3.3.2 Core decoding | In core decoding mode, the output channels (L, R, C, ...) of the Simple Coupling tool are created by processing of the first nSCPL,out processed IMDCT output signals (A'', B'', C'', ...) of the Stereo and Multichannel tool (see Table 21) and assigning them to the output channels. The output channels shall be created as... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.4 Advanced Spectral Extension (A-SPX) post-processing tool | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.4.1 Introduction | The A-SPX post processing tool applies a gain factor of -1,5 dB to the input signal(s) and passes it to the output. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.4.2 Inputs | QinASPX_PP,[0|1|2|...] num_sig complex valued QMF matrices The QinASPX_PP matrices each consist of num_qmf_timeslots columns and num_qmf_subbands rows. If b_5fronts is 1, num_sig is 6, otherwise num_sig is 4. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.4.3 Processing | The decoder shall process the input signals QinASPX_PP,[0|1|2|...] = Q_in_ASPX_PP[i], for i=0,1,2,... to calculate the output signals QoutASPX_PP,[0|1|2|...] = Q_out_ASPX_PP[i], for i=0,1,2,... as specified in Table 26. Table 26: Pseudocode for (i=0; i<num_sig; i++) { for(ts=0; ts<num_qmf_timeslots; ts++) { for(sb=sbx;... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.5 Advanced Coupling (A-CPL) for immersive audio | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.5.1 Introduction | This Advanced Coupling (A-CPL) tool specification extends the specification of the A-CPL tool in ETSI TS 103 190-1 [1], clause 5.7.7 to support A-CPL for immersive channel audio. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.5.2 Processing the immersive_channel_element | When decoding an immersive_channel_element in full decoding mode and immersive_codec_mode ∈ {ASPX_ACPL_1, ASPX_ACPL_2}, the decoder shall utilize the A-CPL tool specified in this clause. In this case the core_channel_config is 7_CH_STATIC and either four or six parallel A- CPL modules are utilized. If b_5front = 1, thi... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6 Advanced Joint Channel Coding (A-JCC) | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.1 Introduction | 5. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 6.1 Introduction | The Advanced Joint Channel Coding (A-JCC) tool improves coding of multiple audio channels. The coding efficiency is achieved by representing the multichannel audio using a five-channel audio signal and parametric side information. The A-JCC tool supports the full decoding mode as specified in clause 5.6.3.5.2 and the c... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.2 Interface | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.2.1 Inputs | QinAJCC,[A|B|C|D|E] five complex valued QMF matrices of five input audio channels to be processed by the A-JCC tool The QinAJCC matrices each consist of num_qmf_timeslots columns and num_qmf_subbands rows. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.2.2 Outputs | QoutAJCC,[L|R|C|…] ajcc_num_out complex valued QMF matrices corresponding to the number of reconstructed audio channels The QoutAJCC matrices each consist of num_qmf_timeslots columns and num_qmf_subbands rows. ajcc_num_out denotes the number of reconstructed output channels and depends on the decoding mode and b_5fron... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.2.3 Controls | The control information for the A-JCC tool consists of decoded and dequantized A-JCC side information. The side information contains parameters to control the dequantization process described in clause 5.6.3.2, the interpolation process described in clause 5.6.3.3, the decorrelation process described in clause 5.6.3.4 ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3 Processing | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.1 Parameter band to QMF subband mapping | The A-JCC parameters are transmitted per parameter band. Like in the A-CPL tool, the parameter bands are groupings of QMF subbands and they have lower frequency resolution than the QMF subbands. The mapping of parameter bands to QMF subbands for the A-JCC tool is the same as the mapping for the A-CPL tool as specified ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.2 Differential decoding and dequantization | To get the quantized values ajcc_<SET>_q from the Huffman decoded values ajcc_<SET>, where <SET> is an identifier for the A-JCC parameter set type and <SET> ∈ {dry1f, dry2f, dry3f, dry4f, dry1b, dry2b, dry3b, dry4b, wet1f, wet2f, wet3f, wet4f, wet5f, wet6f, wet1b, wet2b, wet3b, wet4b, wet5b, wet6b, alpha1, alpha2, beta... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.3 Interpolation | Parameter sets are transmitted either once or twice per frame, determined by the variable ajcc_num_param_sets. Decoded and dequantized A-JCC parameters carried in the bitstream are time-interpolated to calculate values that are applied to the input of the decorrelator and to the ducked output of the decorrelator. Two f... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.4 Decorrelator and transient ducker | The A-JCC processing includes multiple decorrelation processes where ajcc_num_decorr parallel decorrelator instances generate the output signals: Qdecorr_outAJCC,[a|b|…] ajcc_num_decorr complex valued QMF matrices; each one is the output of one of the parallel decorrelator instances The output signals are calculated fr... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.5 Reconstruction of the output channels | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.5.1 Input channels | For the A-JCC reconstruction processing five input channels are present. Their elements are addressed as: First input channel x0(ts,sb) ∈ QinAJCC,A Second input channel x1(ts,sb) ∈ QinAJCC,B Third input channel x2(ts,sb) ∈ QinAJCC,C Fourth input channel x3(ts,sb) ∈ QinAJCC,D Fifth input channel x4(ts,sb) ∈ QinAJCC,E ET... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.5.2 A-JCC full decoding mode | The reconstructed output channels in full decoding mode are addressed as: Left output channel z0(ts,sb) ∈ QoutAJCC,L Right output channel z1(ts,sb) ∈ QoutAJCC,R Centre output channel z2(ts,sb) ∈ QoutAJCC,C Left screen output channel z3(ts,sb) ∈ QoutAJCC,Lscr Right screen output channel z4(ts,sb) ∈ QoutAJCC,Rscr Left su... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.6.3.5.3 A-JCC core decoding mode | The reconstructed output channels in core decoding mode are addressed as: Left output channel z0(ts,sb) ∈ QoutAJCC,L Right output channel z1(ts,sb) ∈ QoutAJCC,R Centre output channel z2(ts,sb) ∈ QoutAJCC,C Left surround output channel z3(ts,sb) ∈ QoutAJCC,Ls Right surround output channel z4(ts,sb) ∈ QoutAJCC,Rs Left to... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7 Advanced Joint Object Coding (A-JOC) | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.1 Introduction | The Advanced Joint Object Coding (A-JOC) tool is a coding tool for improved coding of a multiple number of audio objects. To gain coding efficiency this tool supports the reconstruction of audio objects out of a lower number of joint input audio objects and low overhead side information. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.2 Interface | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.2.1 Inputs | QinAJOC,[a|b|…] mAJOC complex valued QMF matrices for mAJOC objects to be processed by the A-JOC tool The QinAJOC matrices each consist of num_qmf_timeslots columns and num_qmf_subbands rows. mAJOC is the number for num_dmx_signals objects in the coded A-JOC domain. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.2.2 Outputs | QoutAJOC,[a|b|…] nAJOC complex valued QMF matrices corresponding to the number of reconstructed objects The QoutAJOC matrices each consist of num_qmf_timeslots columns and num_qmf_subbands rows. nAJOC is the number for num_umx_signals reconstructed objects. |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.2.3 Controls | The control information for the A-JOC tool consists of decoded and dequantized A-JOC side information. The side information contains parameters to control the dequantization process described in clause 5.7.3.3, the interpolation process described in clause 5.7.3.4, the decorrelation process described in clause 5.7.3.5,... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3 Processing | |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.1 Parameter band to QMF subband mapping | The parameters inside the Advanced Joint Object Coding sideinfo are transmitted per parameter band. The number of parameter bands is coded using the element ajoc_num_bands_code. The value of this element indicates the number of transmitted parameter bands, ajoc_num_bands, as shown in Table 102. The mapping of grouping ... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.2 Differential decoding | The pseudocode in Table 44 describes the process to get quantized values mtx_dry_qo and mtx_wet_qo for A-JOC object o from the Huffman decoded values mix_mtx_dry and mix_mtx_wet, where 0 ≤ o < nAJOC and 0 ≤ ch < mAJOC. Table 44: Pseudocode // differential decoding for A-JOC object o // input: array mix_mtx_dry[o][dp][c... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.3 Dequantization | The dequantized values mtx_dry_dq and mtx_wet_dq are obtained from mtx_dry_q and mtx_wet_q using Table 45 and Table 47 if the quantization mode is set to coarse (ajoc_quant_select=1), and using Table 46 and Table 48 if the quantization mode is set to fine (ajoc_quant_select=0). The dry values are dequantized into range... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.4 Parameter time interpolation | Decoded and dequantized advanced joint object coding parameters carried in the bitstream are time-interpolated for further processing. Each frame can carry one or two new parameter sets. In the case of interpolation over multiple frames, a frame can have no new parameter sets. The bitstream element ajoc_num_dpoints sig... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.5 Decorrelator and transient ducker | This clause specifies function ajoc_decorrelate() as an extension of the decorrelator specified for advanced coupling in ETSI TS 103 190-1 [1], clause 5.7.7.4.2. The A-JOC processing includes multiple decorrelation processes where ajoc_num_decorr parallel decorrelator instances generate the output signals: Qdecorr_outA... |
9cc71134e4c63574779a8d3d5ef00858 | 103 190-2 | 5.7.3.6 Signal reconstruction using matrices |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.