*/
/**
- * @file libavcodec/aac.h
+ * @file
* AAC definitions and structures
* @author Oded Shimon ( ods15 ods15 dyndns org )
* @author Maxim Gavrilov ( maxim.gavrilov gmail com )
#ifndef AVCODEC_AAC_H
#define AVCODEC_AAC_H
-#include "libavutil/internal.h"
#include "avcodec.h"
#include "dsputil.h"
+#include "fft.h"
#include "mpeg4audio.h"
+#include "sbr.h"
#include <stdint.h>
-#define AAC_INIT_VLC_STATIC(num, size) \
- INIT_VLC_STATIC(&vlc_spectral[num], 6, ff_aac_spectral_sizes[num], \
- ff_aac_spectral_bits[num], sizeof( ff_aac_spectral_bits[num][0]), sizeof( ff_aac_spectral_bits[num][0]), \
- ff_aac_spectral_codes[num], sizeof(ff_aac_spectral_codes[num][0]), sizeof(ff_aac_spectral_codes[num][0]), \
- size);
-
#define MAX_CHANNELS 64
#define MAX_ELEM_ID 16
#define TNS_MAX_ORDER 20
-enum AudioObjectType {
- AOT_NULL,
- // Support? Name
- AOT_AAC_MAIN, ///< Y Main
- AOT_AAC_LC, ///< Y Low Complexity
- AOT_AAC_SSR, ///< N (code in SoC repo) Scalable Sample Rate
- AOT_AAC_LTP, ///< N (code in SoC repo) Long Term Prediction
- AOT_SBR, ///< N (in progress) Spectral Band Replication
- AOT_AAC_SCALABLE, ///< N Scalable
- AOT_TWINVQ, ///< N Twin Vector Quantizer
- AOT_CELP, ///< N Code Excited Linear Prediction
- AOT_HVXC, ///< N Harmonic Vector eXcitation Coding
- AOT_TTSI = 12, ///< N Text-To-Speech Interface
- AOT_MAINSYNTH, ///< N Main Synthesis
- AOT_WAVESYNTH, ///< N Wavetable Synthesis
- AOT_MIDI, ///< N General MIDI
- AOT_SAFX, ///< N Algorithmic Synthesis and Audio Effects
- AOT_ER_AAC_LC, ///< N Error Resilient Low Complexity
- AOT_ER_AAC_LTP = 19, ///< N Error Resilient Long Term Prediction
- AOT_ER_AAC_SCALABLE, ///< N Error Resilient Scalable
- AOT_ER_TWINVQ, ///< N Error Resilient Twin Vector Quantizer
- AOT_ER_BSAC, ///< N Error Resilient Bit-Sliced Arithmetic Coding
- AOT_ER_AAC_LD, ///< N Error Resilient Low Delay
- AOT_ER_CELP, ///< N Error Resilient Code Excited Linear Prediction
- AOT_ER_HVXC, ///< N Error Resilient Harmonic Vector eXcitation Coding
- AOT_ER_HILN, ///< N Error Resilient Harmonic and Individual Lines plus Noise
- AOT_ER_PARAM, ///< N Error Resilient Parametric
- AOT_SSC, ///< N SinuSoidal Coding
-};
-
enum RawDataBlockType {
TYPE_SCE,
TYPE_CPE,
AFTER_IMDCT = 3,
};
+/**
+ * Output configuration status
+ */
+enum OCStatus {
+ OC_NONE, //< Output unconfigured
+ OC_TRIAL_PCE, //< Output configuration under trial specified by an inband PCE
+ OC_TRIAL_FRAME, //< Output configuration under trial specified by a frame header
+ OC_GLOBAL_HDR, //< Output configuration set in a global header but not yet locked
+ OC_LOCKED, //< Output configuration locked in place
+};
+
/**
* Predictor State
*/
#define MAX_PREDICTORS 672
+#define SCALE_DIV_512 36 ///< scalefactor difference that corresponds to scale difference in 512 times
+#define SCALE_ONE_POS 140 ///< scalefactor index that corresponds to scale=1.0
+#define SCALE_MAX_POS 255 ///< scalefactor index maximum value
+#define SCALE_MAX_DIFF 60 ///< maximum scalefactor difference allowed by standard
+#define SCALE_DIFF_ZERO 60 ///< codebook index corresponding to zero scalefactor indices difference
+
/**
* Individual Channel Stream
*/
int num_window_groups;
uint8_t group_len[8];
const uint16_t *swb_offset; ///< table of offsets to the lowest spectral coefficient of a scalefactor band, sfb, for a particular window
+ const uint8_t *swb_sizes; ///< table of scalefactor band sizes for a particular window
int num_swb; ///< number of scalefactor window bands
int num_windows;
int tns_max_bands;
typedef struct {
int num_pulse;
+ int start;
int pos[4];
int amp[4];
} Pulse;
typedef struct {
IndividualChannelStream ics;
TemporalNoiseShaping tns;
- enum BandType band_type[120]; ///< band types
+ Pulse pulse;
+ enum BandType band_type[128]; ///< band types
int band_type_run_end[120]; ///< band type run end points
float sf[120]; ///< scalefactors
- DECLARE_ALIGNED_16(float, coeffs[1024]); ///< coefficients for IMDCT
- DECLARE_ALIGNED_16(float, saved[512]); ///< overlap
- DECLARE_ALIGNED_16(float, ret[1024]); ///< PCM output
+ int sf_idx[128]; ///< scalefactor indices (used by encoder)
+ uint8_t zeroes[128]; ///< band is not coded (used by encoder)
+ DECLARE_ALIGNED(16, float, coeffs)[1024]; ///< coefficients for IMDCT
+ DECLARE_ALIGNED(16, float, saved)[1024]; ///< overlap
+ DECLARE_ALIGNED(16, float, ret)[2048]; ///< PCM output
PredictorState predictor_state[MAX_PREDICTORS];
} SingleChannelElement;
*/
typedef struct {
// CPE specific
- uint8_t ms_mask[120]; ///< Set if mid/side stereo is used for each scalefactor window band
+ int common_window; ///< Set if channels share a common 'IndividualChannelStream' in bitstream.
+ int ms_mode; ///< Signals mid/side stereo flags coding mode (used by encoder)
+ uint8_t ms_mask[128]; ///< Set if mid/side stereo is used for each scalefactor window band
// shared
SingleChannelElement ch[2];
// CCE specific
ChannelCoupling coup;
+ SpectralBandReplication sbr;
} ChannelElement;
/**
* main AAC context
*/
typedef struct {
- AVCodecContext * avccontext;
+ AVCodecContext *avctx;
MPEG4AudioConfig m4ac;
enum ChannelPosition che_pos[4][MAX_ELEM_ID]; /**< channel element channel mapping with the
* first index as the first 4 raw data block types
*/
- ChannelElement * che[4][MAX_ELEM_ID];
+ ChannelElement *che[4][MAX_ELEM_ID];
+ ChannelElement *tag_che_map[4][MAX_ELEM_ID];
+ int tags_mapped;
/** @} */
/**
* @defgroup temporary aligned temporary buffers (We do not want to have these on the stack.)
* @{
*/
- DECLARE_ALIGNED_16(float, buf_mdct[1024]);
+ DECLARE_ALIGNED(16, float, buf_mdct)[1024];
/** @} */
/**
* @defgroup tables Computed / set up during initialization.
* @{
*/
- MDCTContext mdct;
- MDCTContext mdct_small;
+ FFTContext mdct;
+ FFTContext mdct_small;
DSPContext dsp;
int random_state;
/** @} */
int sf_offset; ///< offset into pow2sf_tab as appropriate for dsp.float_to_int16
/** @} */
- DECLARE_ALIGNED(16, float, temp[128]);
+ DECLARE_ALIGNED(16, float, temp)[128];
+
+ enum OCStatus output_configured;
} AACContext;
#endif /* AVCODEC_AAC_H */
projects / ffmpeg / blobdiff