2 * Common code between the AC-3 encoder and decoder
3 * Copyright (c) 2000, 2001, 2002 Fabrice Bellard
5 * This file is part of FFmpeg.
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.
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.
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
24 * Common code between the AC-3 encoder and decoder.
30 #define AC3_MAX_CODED_FRAME_SIZE 3840 /* in bytes */
31 #define AC3_MAX_CHANNELS 6 /* including LFE channel */
33 #define AC3_MAX_COEFS 256
34 #define AC3_BLOCK_SIZE 256
35 #define AC3_MAX_BLOCKS 6
36 #define AC3_FRAME_SIZE (AC3_MAX_BLOCKS * 256)
37 #define AC3_WINDOW_SIZE (AC3_BLOCK_SIZE * 2)
38 #define AC3_CRITICAL_BANDS 50
39 #define AC3_MAX_CPL_BANDS 18
41 #include "libavutil/opt.h"
45 /* exponent encoding strategy */
53 /* pre-defined gain values */
54 #define LEVEL_PLUS_3DB 1.4142135623730950
55 #define LEVEL_PLUS_1POINT5DB 1.1892071150027209
56 #define LEVEL_MINUS_1POINT5DB 0.8408964152537145
57 #define LEVEL_MINUS_3DB 0.7071067811865476
58 #define LEVEL_MINUS_4POINT5DB 0.5946035575013605
59 #define LEVEL_MINUS_6DB 0.5000000000000000
60 #define LEVEL_MINUS_9DB 0.3535533905932738
61 #define LEVEL_ZERO 0.0000000000000000
62 #define LEVEL_ONE 1.0000000000000000
64 /** Delta bit allocation strategy */
72 /** Channel mode (audio coding mode) */
74 AC3_CHMODE_DUALMONO = 0,
84 typedef struct AC3BitAllocParameters {
87 int slow_gain, slow_decay, fast_decay, db_per_bit, floor;
88 int cpl_fast_leak, cpl_slow_leak;
89 } AC3BitAllocParameters;
92 * @struct AC3HeaderInfo
93 * Coded AC-3 header values up to the lfeon element, plus derived values.
96 /** @defgroup coded Coded elements
102 uint8_t bitstream_id;
103 uint8_t bitstream_mode;
104 uint8_t channel_mode;
107 int substreamid; ///< substream identification
108 int center_mix_level; ///< Center mix level index
109 int surround_mix_level; ///< Surround mix level index
110 uint16_t channel_map;
111 int num_blocks; ///< number of audio blocks
114 /** @defgroup derived Derived values
118 uint16_t sample_rate;
122 int64_t channel_layout;
127 EAC3_FRAME_TYPE_INDEPENDENT = 0,
128 EAC3_FRAME_TYPE_DEPENDENT,
129 EAC3_FRAME_TYPE_AC3_CONVERT,
130 EAC3_FRAME_TYPE_RESERVED
134 * Encoding Options used by AVOption.
136 typedef struct AC3EncOptions {
137 /* AC-3 metadata options*/
140 float center_mix_level;
141 float surround_mix_level;
142 int dolby_surround_mode;
143 int audio_production_info;
149 int preferred_stereo_downmix;
150 float ltrt_center_mix_level;
151 float ltrt_surround_mix_level;
152 float loro_center_mix_level;
153 float loro_surround_mix_level;
155 int dolby_surround_ex_mode;
156 int dolby_headphone_mode;
157 int ad_converter_type;
159 /* other encoding options */
160 int allow_per_frame_metadata;
164 void ff_ac3_common_init(void);
166 extern const int64_t ff_ac3_channel_layouts[];
167 extern const AVOption ff_ac3_options[];
169 extern AVCodec ff_ac3_float_encoder;
170 extern AVCodec ff_ac3_fixed_encoder;
173 * Calculate the log power-spectral density of the input signal.
174 * This gives a rough estimate of signal power in the frequency domain by using
175 * the spectral envelope (exponents). The psd is also separately grouped
176 * into critical bands for use in the calculating the masking curve.
177 * 128 units in psd = -6 dB. The dbknee parameter in AC3BitAllocParameters
178 * determines the reference level.
180 * @param[in] exp frequency coefficient exponents
181 * @param[in] start starting bin location
182 * @param[in] end ending bin location
183 * @param[out] psd signal power for each frequency bin
184 * @param[out] band_psd signal power for each critical band
186 void ff_ac3_bit_alloc_calc_psd(int8_t *exp, int start, int end, int16_t *psd,
190 * Calculate the masking curve.
191 * First, the excitation is calculated using parameters in s and the signal
192 * power in each critical band. The excitation is compared with a predefined
193 * hearing threshold table to produce the masking curve. If delta bit
194 * allocation information is provided, it is used for adjusting the masking
195 * curve, usually to give a closer match to a better psychoacoustic model.
197 * @param[in] s adjustable bit allocation parameters
198 * @param[in] band_psd signal power for each critical band
199 * @param[in] start starting bin location
200 * @param[in] end ending bin location
201 * @param[in] fast_gain fast gain (estimated signal-to-mask ratio)
202 * @param[in] is_lfe whether or not the channel being processed is the LFE
203 * @param[in] dba_mode delta bit allocation mode (none, reuse, or new)
204 * @param[in] dba_nsegs number of delta segments
205 * @param[in] dba_offsets location offsets for each segment
206 * @param[in] dba_lengths length of each segment
207 * @param[in] dba_values delta bit allocation for each segment
208 * @param[out] mask calculated masking curve
209 * @return returns 0 for success, non-zero for error
211 int ff_ac3_bit_alloc_calc_mask(AC3BitAllocParameters *s, int16_t *band_psd,
212 int start, int end, int fast_gain, int is_lfe,
213 int dba_mode, int dba_nsegs, uint8_t *dba_offsets,
214 uint8_t *dba_lengths, uint8_t *dba_values,
217 #endif /* AVCODEC_AC3_H */