path: root/libavcodec/aacenc.h (plain)
blob: 9d244fd1f58239784e29121004f657aa2ef2b7b5

1	/*
2	* AAC encoder
3	* Copyright (C) 2008 Konstantin Shishkov
4	*
5	* This file is part of FFmpeg.
6	*
7	* FFmpeg is free software; you can redistribute it and/or
8	* modify it under the terms of the GNU Lesser General Public
9	* License as published by the Free Software Foundation; either
10	* version 2.1 of the License, or (at your option) any later version.
11	*
12	* FFmpeg is distributed in the hope that it will be useful,
13	* but WITHOUT ANY WARRANTY; without even the implied warranty of
14	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15	* Lesser General Public License for more details.
16	*
17	* You should have received a copy of the GNU Lesser General Public
18	* License along with FFmpeg; if not, write to the Free Software
19	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20	*/
21
22	#ifndef AVCODEC_AACENC_H
23	#define AVCODEC_AACENC_H
24
25	#include "libavutil/float_dsp.h"
26	#include "avcodec.h"
27	#include "put_bits.h"
28
29	#include "aac.h"
30	#include "audio_frame_queue.h"
31	#include "psymodel.h"
32
33	#include "lpc.h"
34
35	typedef enum AACCoder {
36	AAC_CODER_ANMR = 0,
37	AAC_CODER_TWOLOOP,
38	AAC_CODER_FAST,
39
40	AAC_CODER_NB,
41	}AACCoder;
42
43	typedef struct AACEncOptions {
44	int coder;
45	int pns;
46	int tns;
47	int ltp;
48	int pred;
49	int mid_side;
50	int intensity_stereo;
51	} AACEncOptions;
52
53	struct AACEncContext;
54
55	typedef struct AACCoefficientsEncoder {
56	void (*search_for_quantizers)(AVCodecContext *avctx, struct AACEncContext *s,
57	SingleChannelElement *sce, const float lambda);
58	void (*encode_window_bands_info)(struct AACEncContext *s, SingleChannelElement *sce,
59	int win, int group_len, const float lambda);
60	void (*quantize_and_encode_band)(struct AACEncContext *s, PutBitContext *pb, const float *in, float *out, int size,
61	int scale_idx, int cb, const float lambda, int rtz);
62	void (*encode_tns_info)(struct AACEncContext *s, SingleChannelElement *sce);
63	void (*encode_ltp_info)(struct AACEncContext *s, SingleChannelElement *sce, int common_window);
64	void (*encode_main_pred)(struct AACEncContext *s, SingleChannelElement *sce);
65	void (*adjust_common_pred)(struct AACEncContext *s, ChannelElement *cpe);
66	void (*adjust_common_ltp)(struct AACEncContext *s, ChannelElement *cpe);
67	void (*apply_main_pred)(struct AACEncContext *s, SingleChannelElement *sce);
68	void (*apply_tns_filt)(struct AACEncContext *s, SingleChannelElement *sce);
69	void (*update_ltp)(struct AACEncContext *s, SingleChannelElement *sce);
70	void (*ltp_insert_new_frame)(struct AACEncContext *s);
71	void (*set_special_band_scalefactors)(struct AACEncContext *s, SingleChannelElement *sce);
72	void (*search_for_pns)(struct AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce);
73	void (*mark_pns)(struct AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce);
74	void (*search_for_tns)(struct AACEncContext *s, SingleChannelElement *sce);
75	void (*search_for_ltp)(struct AACEncContext *s, SingleChannelElement *sce, int common_window);
76	void (*search_for_ms)(struct AACEncContext *s, ChannelElement *cpe);
77	void (*search_for_is)(struct AACEncContext *s, AVCodecContext *avctx, ChannelElement *cpe);
78	void (*search_for_pred)(struct AACEncContext *s, SingleChannelElement *sce);
79	} AACCoefficientsEncoder;
80
81	extern AACCoefficientsEncoder ff_aac_coders[];
82
83	typedef struct AACQuantizeBandCostCacheEntry {
84	float rd;
85	float energy;
86	int bits;
87	char cb;
88	char rtz;
89	uint16_t generation;
90	} AACQuantizeBandCostCacheEntry;
91
92	/**
93	* AAC encoder context
94	*/
95	typedef struct AACEncContext {
96	AVClass *av_class;
97	AACEncOptions options; ///< encoding options
98	PutBitContext pb;
99	FFTContext mdct1024; ///< long (1024 samples) frame transform context
100	FFTContext mdct128; ///< short (128 samples) frame transform context
101	AVFloatDSPContext *fdsp;
102	float *planar_samples[8]; ///< saved preprocessed input
103
104	int profile; ///< copied from avctx
105	LPCContext lpc; ///< used by TNS
106	int samplerate_index; ///< MPEG-4 samplerate index
107	int channels; ///< channel count
108	const uint8_t *chan_map; ///< channel configuration map
109
110	ChannelElement *cpe; ///< channel elements
111	FFPsyContext psy;
112	struct FFPsyPreprocessContext* psypp;
113	AACCoefficientsEncoder *coder;
114	int cur_channel; ///< current channel for coder context
115	int random_state;
116	float lambda;
117	int last_frame_pb_count; ///< number of bits for the previous frame
118	float lambda_sum; ///< sum(lambda), for Qvg reporting
119	int lambda_count; ///< count(lambda), for Qvg reporting
120	enum RawDataBlockType cur_type; ///< channel group type cur_channel belongs to
121
122	AudioFrameQueue afq;
123	DECLARE_ALIGNED(16, int, qcoefs)[96]; ///< quantized coefficients
124	DECLARE_ALIGNED(32, float, scoefs)[1024]; ///< scaled coefficients
125
126	uint16_t quantize_band_cost_cache_generation;
127	AACQuantizeBandCostCacheEntry quantize_band_cost_cache[256][128]; ///< memoization area for quantize_band_cost
128
129	void (*abs_pow34)(float *out, const float *in, const int size);
130	void (*quant_bands)(int *out, const float *in, const float *scaled,
131	int size, int is_signed, int maxval, const float Q34,
132	const float rounding);
133
134	struct {
135	float *samples;
136	} buffer;
137	} AACEncContext;
138
139	void ff_aac_dsp_init_x86(AACEncContext *s);
140	void ff_aac_coder_init_mips(AACEncContext *c);
141	void ff_quantize_band_cost_cache_init(struct AACEncContext *s);
142
143
144	#endif /* AVCODEC_AACENC_H */
145