2 * AAC definitions and structures
3 * Copyright (c) 2005-2006 Oded Shimon ( ods15 ods15 dyndns org )
4 * Copyright (c) 2006-2007 Maxim Gavrilov ( maxim.gavrilov gmail com )
6 * This file is part of FFmpeg.
8 * FFmpeg is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
13 * FFmpeg is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 * Lesser General Public License for more details.
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with FFmpeg; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
24 * @file libavcodec/aac.h
25 * AAC definitions and structures
26 * @author Oded Shimon ( ods15 ods15 dyndns org )
27 * @author Maxim Gavrilov ( maxim.gavrilov gmail com )
35 #include "mpeg4audio.h"
39 #define AAC_INIT_VLC_STATIC(num, size) \
40 INIT_VLC_STATIC(&vlc_spectral[num], 6, ff_aac_spectral_sizes[num], \
41 ff_aac_spectral_bits[num], sizeof( ff_aac_spectral_bits[num][0]), sizeof( ff_aac_spectral_bits[num][0]), \
42 ff_aac_spectral_codes[num], sizeof(ff_aac_spectral_codes[num][0]), sizeof(ff_aac_spectral_codes[num][0]), \
45 #define MAX_CHANNELS 64
46 #define MAX_ELEM_ID 16
48 #define TNS_MAX_ORDER 20
50 enum RawDataBlockType {
61 enum ExtensionPayloadID {
65 EXT_DYNAMIC_RANGE = 0xb,
67 EXT_SBR_DATA_CRC = 0xe,
78 ZERO_BT = 0, ///< Scalefactors and spectral data are all zero.
79 FIRST_PAIR_BT = 5, ///< This and later band types encode two values (rather than four) with one code word.
80 ESC_BT = 11, ///< Spectral data are coded with an escape sequence.
81 NOISE_BT = 13, ///< Spectral data are scaled white noise not coded in the bitstream.
82 INTENSITY_BT2 = 14, ///< Scalefactor data are intensity stereo positions.
83 INTENSITY_BT = 15, ///< Scalefactor data are intensity stereo positions.
86 #define IS_CODEBOOK_UNSIGNED(x) ((x - 1) & 10)
88 enum ChannelPosition {
89 AAC_CHANNEL_FRONT = 1,
97 * The point during decoding at which channel coupling is applied.
101 BETWEEN_TNS_AND_IMDCT,
117 #define MAX_PREDICTORS 672
119 #define SCALE_DIV_512 36 ///< scalefactor difference that corresponds to scale difference in 512 times
120 #define SCALE_ONE_POS 140 ///< scalefactor index that corresponds to scale=1.0
121 #define SCALE_MAX_POS 255 ///< scalefactor index maximum value
122 #define SCALE_MAX_DIFF 60 ///< maximum scalefactor difference allowed by standard
123 #define SCALE_DIFF_ZERO 60 ///< codebook index corresponding to zero scalefactor indices difference
126 * Individual Channel Stream
129 uint8_t max_sfb; ///< number of scalefactor bands per group
130 enum WindowSequence window_sequence[2];
131 uint8_t use_kb_window[2]; ///< If set, use Kaiser-Bessel window, otherwise use a sinus window.
132 int num_window_groups;
133 uint8_t group_len[8];
134 const uint16_t *swb_offset; ///< table of offsets to the lowest spectral coefficient of a scalefactor band, sfb, for a particular window
135 const uint8_t *swb_sizes; ///< table of scalefactor band sizes for a particular window
136 int num_swb; ///< number of scalefactor window bands
139 int predictor_present;
140 int predictor_initialized;
141 int predictor_reset_group;
142 uint8_t prediction_used[41];
143 } IndividualChannelStream;
146 * Temporal Noise Shaping
154 float coef[8][4][TNS_MAX_ORDER];
155 } TemporalNoiseShaping;
158 * Dynamic Range Control - decoded from the bitstream but not processed further.
161 int pce_instance_tag; ///< Indicates with which program the DRC info is associated.
162 int dyn_rng_sgn[17]; ///< DRC sign information; 0 - positive, 1 - negative
163 int dyn_rng_ctl[17]; ///< DRC magnitude information
164 int exclude_mask[MAX_CHANNELS]; ///< Channels to be excluded from DRC processing.
165 int band_incr; ///< Number of DRC bands greater than 1 having DRC info.
166 int interpolation_scheme; ///< Indicates the interpolation scheme used in the SBR QMF domain.
167 int band_top[17]; ///< Indicates the top of the i-th DRC band in units of 4 spectral lines.
168 int prog_ref_level; /**< A reference level for the long-term program audio level for all
171 } DynamicRangeControl;
181 * coupling parameters
184 enum CouplingPoint coupling_point; ///< The point during decoding at which coupling is applied.
185 int num_coupled; ///< number of target elements
186 enum RawDataBlockType type[8]; ///< Type of channel element to be coupled - SCE or CPE.
187 int id_select[8]; ///< element id
188 int ch_select[8]; /**< [0] shared list of gains; [1] list of gains for right channel;
189 * [2] list of gains for left channel; [3] lists of gains for both channels
195 * Single Channel Element - used for both SCE and LFE elements.
198 IndividualChannelStream ics;
199 TemporalNoiseShaping tns;
201 enum BandType band_type[128]; ///< band types
202 int band_type_run_end[120]; ///< band type run end points
203 float sf[120]; ///< scalefactors
204 int sf_idx[128]; ///< scalefactor indices (used by encoder)
205 uint8_t zeroes[128]; ///< band is not coded (used by encoder)
206 DECLARE_ALIGNED_16(float, coeffs[1024]); ///< coefficients for IMDCT
207 DECLARE_ALIGNED_16(float, saved[1024]); ///< overlap
208 DECLARE_ALIGNED_16(float, ret[1024]); ///< PCM output
209 PredictorState predictor_state[MAX_PREDICTORS];
210 } SingleChannelElement;
213 * channel element - generic struct for SCE/CPE/CCE/LFE
217 int common_window; ///< Set if channels share a common 'IndividualChannelStream' in bitstream.
218 int ms_mode; ///< Signals mid/side stereo flags coding mode (used by encoder)
219 uint8_t ms_mask[128]; ///< Set if mid/side stereo is used for each scalefactor window band
221 SingleChannelElement ch[2];
223 ChannelCoupling coup;
230 AVCodecContext * avccontext;
232 MPEG4AudioConfig m4ac;
234 int is_saved; ///< Set if elements have stored overlap from previous frame.
235 DynamicRangeControl che_drc;
238 * @defgroup elements Channel element related data.
241 enum ChannelPosition che_pos[4][MAX_ELEM_ID]; /**< channel element channel mapping with the
242 * first index as the first 4 raw data block types
244 ChannelElement * che[4][MAX_ELEM_ID];
245 ChannelElement * tag_che_map[4][MAX_ELEM_ID];
250 * @defgroup temporary aligned temporary buffers (We do not want to have these on the stack.)
253 DECLARE_ALIGNED_16(float, buf_mdct[1024]);
257 * @defgroup tables Computed / set up during initialization.
261 MDCTContext mdct_small;
267 * @defgroup output Members used for output interleaving.
270 float *output_data[MAX_CHANNELS]; ///< Points to each element's 'ret' buffer (PCM output).
271 float add_bias; ///< offset for dsp.float_to_int16
272 float sf_scale; ///< Pre-scale for correct IMDCT and dsp.float_to_int16.
273 int sf_offset; ///< offset into pow2sf_tab as appropriate for dsp.float_to_int16
276 DECLARE_ALIGNED(16, float, temp[128]);
278 int output_configured;
281 #endif /* AVCODEC_AAC_H */