1 /* 2 * TwoLAME: an optimized MPEG Audio Layer Two encoder 3 * 4 * Copyright (C) 2001-2004 Michael Cheng 5 * Copyright (C) 2004-2006 The TwoLAME Project 6 * 7 * This library 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 * This library 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 this library; if not, write to the Free Software 19 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 20 * 21 * $Id$ 22 * 23 */ 24 25 26 #ifndef _COMMON_H 27 #define _COMMON_H 28 29 #ifdef _WIN32 30 # include "../win32/configwin.h" 31 #else 32 # include "config.h" 33 #endif 34 35 #include "twolame.h" 36 37 38 39 /*************************************************************************************** 40 Common Definitions 41 ****************************************************************************************/ 42 43 #ifndef FLOAT 44 #define FLOAT double 45 #endif 46 47 #define NULL_CHAR '\0' 48 49 #define MAX_U_32_NUM 0xFFFFFFFF 50 #ifndef PI 51 #define PI 3.14159265358979 52 #endif 53 #ifndef E 54 #define E 2.71828182845 55 #endif 56 #define PI2 PI/2 57 #define PI4 PI/4 58 #define PI64 PI/64 59 #define LN_TO_LOG10 0.2302585093 60 61 #define BITS_IN_A_BYTE 8 62 #define WORD 16 63 #define MAX_NAME_SIZE 255 64 #define SBLIMIT 32 65 #define SSLIMIT 18 66 #define FFT_SIZE 1024 67 #define HAN_SIZE 512 68 #define SCALE_BLOCK 12 69 #define SCALE_RANGE 64 70 #define SCALE 32768 71 #define CRC16_POLYNOMIAL 0x8005 72 #define CRC8_POLYNOMIAL 0x1D 73 74 #define MIN(A, B) ((A) < (B) ? (A) : (B)) 75 #define MAX(A, B) ((A) > (B) ? (A) : (B)) 76 77 78 /* This is the smallest MNR a subband can have before it is counted 79 as 'noisy' by the logic which chooses the number of JS subbands */ 80 81 #define NOISY_MIN_MNR 0.0 82 83 84 /*************************************************************************************** 85 Psychacoustic Model 1/3 Definitions 86 ****************************************************************************************/ 87 88 #define NOISE 10 89 #define TONE 20 90 #define DBMIN -200.0 91 #define LAST -1 92 #define STOP -100 93 #define POWERNORM 90.3090 /* = 20 * log10(32768) to normalize */ 94 /* max output power to 96 dB per spec */ 95 96 97 /*************************************************************************************** 98 Psychoacoustic Model 2/4 Definitions 99 ****************************************************************************************/ 100 #define LXMIN 32.0 101 102 103 104 /*************************************************************************************** 105 psycho 0 mem struct 106 ****************************************************************************************/ 107 108 typedef struct psycho_0_mem_struct { 109 FLOAT ath_min[SBLIMIT]; 110 } psycho_0_mem; 111 112 113 114 /*************************************************************************************** 115 psycho 1 mem struct 116 ****************************************************************************************/ 117 #define DBTAB 1000 118 119 typedef struct { 120 int line; 121 FLOAT bark, hear, x; 122 } g_thres, *g_ptr; 123 124 typedef struct { 125 FLOAT x; 126 int type, next, map; 127 } mask, *mask_ptr; 128 129 typedef struct psycho_1_mem_struct { 130 int off[2]; 131 FLOAT fft_buf[2][1408]; 132 int *cbound; 133 int crit_band; 134 int sub_size; 135 mask_ptr power; 136 g_ptr ltg; 137 FLOAT dbtable[DBTAB]; 138 } psycho_1_mem; 139 140 141 142 /*************************************************************************************** 143 Psycho3 memory structure 144 ****************************************************************************************/ 145 #define HBLKSIZE 513 146 147 #define SUBSIZE 136 148 typedef struct psycho_3_mem_struct { 149 int off[2]; 150 int freq_subset[SUBSIZE]; 151 FLOAT bark[HBLKSIZE]; 152 FLOAT ath[HBLKSIZE]; 153 FLOAT fft_buf[2][1408]; 154 #define CRITBANDMAX 32 /* this is much higher than it needs to be. really only about 24 */ 155 int cbands; /* How many critical bands there really are */ 156 int cbandindex[CRITBANDMAX]; /* The spectral line index of the start of each critical band */ 157 FLOAT dbtable[DBTAB]; 158 } psycho_3_mem; 159 160 161 162 /*************************************************************************************** 163 Psycho2 & 4 memory structure 164 ****************************************************************************************/ 165 166 #define LOGBLKSIZE 10 167 #define BLKSIZE 1024 168 #define HBLKSIZE 513 169 #define CBANDS 64 170 #define TRIGTABLESIZE 6284 171 #define TRIGTABLESCALE 2000.0 172 typedef int ICB[CBANDS]; 173 typedef int IHBLK[HBLKSIZE]; 174 typedef FLOAT F32[32]; 175 typedef FLOAT F2_32[2][32]; 176 typedef FLOAT FCB[CBANDS]; 177 typedef FLOAT FCBCB[CBANDS][CBANDS]; 178 typedef FLOAT FBLK[BLKSIZE]; 179 typedef FLOAT FHBLK[HBLKSIZE]; 180 typedef FLOAT F2HBLK[2][HBLKSIZE]; 181 typedef FLOAT F22HBLK[2][2][HBLKSIZE]; 182 typedef FLOAT DCB[CBANDS]; 183 184 typedef struct psycho_4_mem_struct { 185 int new; 186 int old; 187 int oldest; 188 189 int flush; 190 int sync_flush; 191 int syncsize; 192 193 FLOAT grouped_c[CBANDS]; 194 FLOAT grouped_e[CBANDS]; 195 FLOAT nb[CBANDS]; 196 FLOAT cb[CBANDS]; 197 FLOAT tb[CBANDS]; 198 FLOAT ecb[CBANDS]; 199 FLOAT bc[CBANDS]; 200 FLOAT cbval[CBANDS]; 201 FLOAT rnorm[CBANDS]; 202 FLOAT wsamp_r[BLKSIZE], phi[BLKSIZE], energy[BLKSIZE], window[BLKSIZE]; 203 FLOAT ath[HBLKSIZE], thr[HBLKSIZE], c[HBLKSIZE]; 204 FLOAT fthr[HBLKSIZE], absthr[HBLKSIZE]; // psy2 only 205 int numlines[CBANDS]; 206 int partition[HBLKSIZE]; 207 FLOAT *tmn; 208 FCB *s; 209 FHBLK *lthr; 210 F2HBLK *r, *phi_sav; 211 FLOAT snrtmp[2][32]; 212 FLOAT cos_table[TRIGTABLESIZE]; 213 } psycho_4_mem, psycho_2_mem; 214 215 216 /*************************************************************************************** 217 Subband utility structures 218 ****************************************************************************************/ 219 220 typedef struct subband_mem_struct { 221 FLOAT x[2][512]; 222 FLOAT m[16][32]; 223 int off[2]; 224 int half[2]; 225 } subband_mem; 226 227 228 229 /*************************************************************************************** 230 Header and frame information 231 ****************************************************************************************/ 232 233 /* Raw Header Information Structure */ 234 typedef struct { 235 int version; 236 int lay; 237 int error_protection; 238 int bitrate_index; 239 int samplerate_idx; 240 int padding; 241 int private_bit; 242 int mode; 243 int mode_ext; 244 int copyright; 245 int original; 246 int emphasis; 247 } frame_header; 248 249 250 251 typedef unsigned int subband_t[2][3][SCALE_BLOCK][SBLIMIT]; 252 typedef FLOAT jsb_sample_t[3][SCALE_BLOCK][SBLIMIT]; 253 typedef FLOAT sb_sample_t[2][3][SCALE_BLOCK][SBLIMIT]; 254 255 256 257 /*************************************************************************************** 258 twolame Global Options structure. 259 Defaults shown in [] 260 ++ means it is an advanced option. Only use it if you know what you're doing. 261 ****************************************************************************************/ 262 struct twolame_options_struct { 263 // Input PCM audio File Information 264 int samplerate_in; // mpeg1: 32000 [44100] 48000 265 // mpeg2: 16000 22050 24000 266 int samplerate_out; 267 int num_channels_in; // Number of channels on the input stream 268 int num_channels_out; // Number of channels on the output stream 269 270 // Output MP2 File Information 271 TWOLAME_MPEG_version version; // 0 mpeg2 [1] mpeg1 272 int bitrate; // for mpeg1:32, 48, 56, 64, 80, 96,112,128,160,[192], 224, 256, 273 // 320, 384 274 // for mpeg2: 8, 16, 24, 32, 40, 48, 56, 64, 80, [96], 112, 128, 144, 160 275 TWOLAME_MPEG_mode mode; 276 TWOLAME_Padding padding; // [PAD_NO] 277 int do_energy_levels; // Write energy level information into the end of the frame [FALSE] 278 int num_ancillary_bits; // Number of reserved ancillary bits [0] (Currently only available 279 // for non-VBR modes) 280 281 // Psychoacoustic Model options 282 int psymodel; // -1, 0, 1, 2, [3], 4 Psy model number 283 FLOAT athlevel; // Adjust the Absolute Threshold of Hearing curve by [0] dB 284 int quickmode; // Only calculate psy model ever X frames [FALSE] 285 int quickcount; // Only calculate psy model every [10] frames 286 287 // VBR Options 288 int vbr; // turn on VBR mode TRUE [FALSE] 289 int vbr_upper_index; // ++ [0] means no upper bitrate set for VBR mode. valid 1-15 290 // depending on mode 291 int vbr_max_bitrate; 292 FLOAT vbrlevel; // Set VBR quality. [0.0] (sensible range -10.0 -> 10.0) 293 294 // Miscellaneous Options That Nobody Ever Uses 295 TWOLAME_Emphasis emphasis; // [n]one, 5(50/15 microseconds), c(ccitt j.17) 296 int copyright; // [FALSE] 297 int original; // [FALSE] 298 int private_bit; // [0] Your very own bit in the header. 299 int error_protection; // [FALSE] 300 301 // Digital Audio Broadcasting Extensions 302 unsigned int do_dab; // Allocate space for the DigitalAudioBroadcasting info [FALSE] 303 unsigned int dab_crc_len; // Number of CRC bytes for DAB [2], 4 304 unsigned int dab_crc[4]; // DAB CRC bytes are inserted here. User must insert them in frame 305 unsigned int dab_xpad_len; // Number of bytes in the XPAD 306 307 // Processing Options 308 int verbosity; // Verbosity of output 0(never output a thing) [2] 100(output 309 // everything) 310 311 312 // Scaling 313 FLOAT scale; 314 FLOAT scale_left; 315 FLOAT scale_right; 316 317 318 319 // Bit allocation stuff 320 int lower_index; 321 int upper_index; 322 int bitrateindextobits[15]; 323 int vbr_frame_count; // Used for debugging VBR 324 325 326 // Used by twolame_encode_frame 327 int twolame_init; 328 short int buffer[2][TWOLAME_SAMPLES_PER_FRAME]; // Sample buffer 329 unsigned int samples_in_buffer; // Number of samples currently in buffer 330 unsigned int psycount; 331 unsigned int num_crc_bits; // Number of bits CRC is calculated on 332 333 unsigned int bit_alloc[2][SBLIMIT]; 334 unsigned int scfsi[2][SBLIMIT]; 335 unsigned int scalar[2][3][SBLIMIT]; 336 unsigned int j_scale[3][SBLIMIT]; 337 FLOAT smrdef[2][32]; 338 FLOAT smr[2][SBLIMIT]; 339 FLOAT max_sc[2][SBLIMIT]; 340 341 subband_t *subband; 342 jsb_sample_t *j_sample; 343 sb_sample_t *sb_sample; 344 345 346 347 /* Resampling stuff */ 348 FLOAT resample_ratio; 349 void *resample_handle[2]; 350 351 352 // memory for psycho models 353 psycho_0_mem *p0mem; 354 psycho_1_mem *p1mem; 355 psycho_2_mem *p2mem; 356 psycho_3_mem *p3mem; 357 psycho_4_mem *p4mem; 358 359 360 // memory for subband 361 subband_mem smem; 362 363 // Frame info 364 frame_header header; 365 int jsbound; // first band of joint stereo coding 366 int sblimit; // total number of sub bands 367 int tablenum; 368 369 int vbrstats[15]; 370 }; 371 372 #endif // _COMMON_H 373 374 375 // vim:ts=4:sw=4:nowrap: 376