Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
tree: 1e9caf9c92
Fetching contributors…

Octocat-spinner-32-eaf2f5

Cannot retrieve contributors at this time

file 611 lines (523 sloc) 23.269 kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610
/*
* G.729, G729 Annex D postfilter
* Copyright (c) 2008 Vladimir Voroshilov
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <inttypes.h>
#include <limits.h>

#include "avcodec.h"
#include "g729.h"
#include "acelp_pitch_delay.h"
#include "g729postfilter.h"
#include "celp_math.h"
#include "acelp_filters.h"
#include "acelp_vectors.h"
#include "celp_filters.h"

#define FRAC_BITS 15
#include "mathops.h"

/**
* short interpolation filter (of length 33, according to spec)
* for computing signal with non-integer delay
*/
static const int16_t ff_g729_interp_filt_short[(ANALYZED_FRAC_DELAYS+1)*SHORT_INT_FILT_LEN] = {
      0, 31650, 28469, 23705, 18050, 12266, 7041, 2873,
      0, -1597, -2147, -1992, -1492, -933, -484, -188,
};

/**
* long interpolation filter (of length 129, according to spec)
* for computing signal with non-integer delay
*/
static const int16_t ff_g729_interp_filt_long[(ANALYZED_FRAC_DELAYS+1)*LONG_INT_FILT_LEN] = {
   0, 31915, 29436, 25569, 20676, 15206, 9639, 4439,
   0, -3390, -5579, -6549, -6414, -5392, -3773, -1874,
   0, 1595, 2727, 3303, 3319, 2850, 2030, 1023,
   0, -887, -1527, -1860, -1876, -1614, -1150, -579,
   0, 501, 859, 1041, 1044, 892, 631, 315,
   0, -266, -453, -543, -538, -455, -317, -156,
   0, 130, 218, 258, 253, 212, 147, 72,
   0, -59, -101, -122, -123, -106, -77, -40,
};

/**
* formant_pp_factor_num_pow[i] = FORMANT_PP_FACTOR_NUM^(i+1)
*/
static const int16_t formant_pp_factor_num_pow[10]= {
  /* (0.15) */
  18022, 9912, 5451, 2998, 1649, 907, 499, 274, 151, 83
};

/**
* formant_pp_factor_den_pow[i] = FORMANT_PP_FACTOR_DEN^(i+1)
*/
static const int16_t formant_pp_factor_den_pow[10] = {
  /* (0.15) */
  22938, 16057, 11240, 7868, 5508, 3856, 2699, 1889, 1322, 925
};

/**
* \brief Residual signal calculation (4.2.1 if G.729)
* \param out [out] output data filtered through A(z/FORMANT_PP_FACTOR_NUM)
* \param filter_coeffs (3.12) A(z/FORMANT_PP_FACTOR_NUM) filter coefficients
* \param in input speech data to process
* \param subframe_size size of one subframe
*
* \note in buffer must contain 10 items of previous speech data before top of the buffer
* \remark It is safe to pass the same buffer for input and output.
*/
static void residual_filter(int16_t* out, const int16_t* filter_coeffs, const int16_t* in,
                            int subframe_size)
{
    int i, n;

    for (n = subframe_size - 1; n >= 0; n--) {
        int sum = 0x800;
        for (i = 0; i < 10; i++)
            sum += filter_coeffs[i] * in[n - i - 1];

        out[n] = in[n] + (sum >> 12);
    }
}

/**
* \brief long-term postfilter (4.2.1)
* \param dsp initialized DSP context
* \param pitch_delay_int integer part of the pitch delay in the first subframe
* \param residual filtering input data
* \param residual_filt [out] speech signal with applied A(z/FORMANT_PP_FACTOR_NUM) filter
* \param subframe_size size of subframe
*
* \return 0 if long-term prediction gain is less than 3dB, 1 - otherwise
*/
static int16_t long_term_filter(DSPContext *dsp, int pitch_delay_int,
                                const int16_t* residual, int16_t *residual_filt,
                                int subframe_size)
{
    int i, k, tmp, tmp2;
    int sum;
    int L_temp0;
    int L_temp1;
    int64_t L64_temp0;
    int64_t L64_temp1;
    int16_t shift;
    int corr_int_num, corr_int_den;

    int ener;
    int16_t sh_ener;

    int16_t gain_num,gain_den; //selected signal's gain numerator and denominator
    int16_t sh_gain_num, sh_gain_den;
    int gain_num_square;

    int16_t gain_long_num,gain_long_den; //filtered through long interpolation filter signal's gain numerator and denominator
    int16_t sh_gain_long_num, sh_gain_long_den;

    int16_t best_delay_int, best_delay_frac;

    int16_t delayed_signal_offset;
    int lt_filt_factor_a, lt_filt_factor_b;

    int16_t * selected_signal;
    const int16_t * selected_signal_const; //Necessary to avoid compiler warning

    int16_t sig_scaled[SUBFRAME_SIZE + RES_PREV_DATA_SIZE];
    int16_t delayed_signal[ANALYZED_FRAC_DELAYS][SUBFRAME_SIZE+1];
    int corr_den[ANALYZED_FRAC_DELAYS][2];

    tmp = 0;
    for(i=0; i<subframe_size + RES_PREV_DATA_SIZE; i++)
        tmp |= FFABS(residual[i]);

    if(!tmp)
        shift = 3;
    else
        shift = av_log2(tmp) - 11;

    if (shift > 0)
        for (i = 0; i < subframe_size + RES_PREV_DATA_SIZE; i++)
            sig_scaled[i] = residual[i] >> shift;
    else
        for (i = 0; i < subframe_size + RES_PREV_DATA_SIZE; i++)
            sig_scaled[i] = residual[i] << -shift;

    /* Start of best delay searching code */
    gain_num = 0;

    ener = dsp->scalarproduct_int16(sig_scaled + RES_PREV_DATA_SIZE,
                                    sig_scaled + RES_PREV_DATA_SIZE,
                                    subframe_size);
    if (ener) {
        sh_ener = FFMAX(av_log2(ener) - 14, 0);
        ener >>= sh_ener;
        /* Search for best pitch delay.

sum{ r(n) * r(k,n) ] }^2
R'(k)^2 := -------------------------
sum{ r(k,n) * r(k,n) }


R(T) := sum{ r(n) * r(n-T) ] }


where
r(n-T) is integer delayed signal with delay T
r(k,n) is non-integer delayed signal with integer delay best_delay
and fractional delay k */

        /* Find integer delay best_delay which maximizes correlation R(T).

This is also equals to numerator of R'(0),
since the fine search (second step) is done with 1/8
precision around best_delay. */
        corr_int_num = 0;
        best_delay_int = pitch_delay_int - 1;
        for (i = pitch_delay_int - 1; i <= pitch_delay_int + 1; i++) {
            sum = dsp->scalarproduct_int16(sig_scaled + RES_PREV_DATA_SIZE,
                                           sig_scaled + RES_PREV_DATA_SIZE - i,
                                           subframe_size);
            if (sum > corr_int_num) {
                corr_int_num = sum;
                best_delay_int = i;
            }
        }
        if (corr_int_num) {
            /* Compute denominator of pseudo-normalized correlation R'(0). */
            corr_int_den = dsp->scalarproduct_int16(sig_scaled - best_delay_int + RES_PREV_DATA_SIZE,
                                                    sig_scaled - best_delay_int + RES_PREV_DATA_SIZE,
                                                    subframe_size);

            /* Compute signals with non-integer delay k (with 1/8 precision),
where k is in [0;6] range.
Entire delay is qual to best_delay+(k+1)/8
This is archieved by applying an interpolation filter of
legth 33 to source signal. */
            for (k = 0; k < ANALYZED_FRAC_DELAYS; k++) {
                ff_acelp_interpolate(&delayed_signal[k][0],
                                     &sig_scaled[RES_PREV_DATA_SIZE - best_delay_int],
                                     ff_g729_interp_filt_short,
                                     ANALYZED_FRAC_DELAYS+1,
                                     8 - k - 1,
                                     SHORT_INT_FILT_LEN,
                                     subframe_size + 1);
            }

            /* Compute denominator of pseudo-normalized correlation R'(k).

corr_den[k][0] is square root of R'(k) denominator, for int(T) == int(T0)
corr_den[k][1] is square root of R'(k) denominator, for int(T) == int(T0)+1

Also compute maximum value of above denominators over all k. */
            tmp = corr_int_den;
            for (k = 0; k < ANALYZED_FRAC_DELAYS; k++) {
                sum = dsp->scalarproduct_int16(&delayed_signal[k][1],
                                               &delayed_signal[k][1],
                                               subframe_size - 1);
                corr_den[k][0] = sum + delayed_signal[k][0 ] * delayed_signal[k][0 ];
                corr_den[k][1] = sum + delayed_signal[k][subframe_size] * delayed_signal[k][subframe_size];

                tmp = FFMAX3(tmp, corr_den[k][0], corr_den[k][1]);
            }

            sh_gain_den = av_log2(tmp) - 14;
            if (sh_gain_den >= 0) {

                sh_gain_num = FFMAX(sh_gain_den, sh_ener);
                /* Loop through all k and find delay that maximizes
R'(k) correlation.
Search is done in [int(T0)-1; intT(0)+1] range
with 1/8 precision. */
                delayed_signal_offset = 1;
                best_delay_frac = 0;
                gain_den = corr_int_den >> sh_gain_den;
                gain_num = corr_int_num >> sh_gain_num;
                gain_num_square = gain_num * gain_num;
                for (k = 0; k < ANALYZED_FRAC_DELAYS; k++) {
                    for (i = 0; i < 2; i++) {
                        int16_t gain_num_short, gain_den_short;
                        int gain_num_short_square;
                        /* Compute numerator of pseudo-normalized
correlation R'(k). */
                        sum = dsp->scalarproduct_int16(&delayed_signal[k][i],
                                                       sig_scaled + RES_PREV_DATA_SIZE,
                                                       subframe_size);
                        gain_num_short = FFMAX(sum >> sh_gain_num, 0);

                        /*
gain_num_short_square gain_num_square
R'(T)^2 = -----------------------, max R'(T)^2= --------------
den gain_den
*/
                        gain_num_short_square = gain_num_short * gain_num_short;
                        gain_den_short = corr_den[k][i] >> sh_gain_den;

                        tmp = MULL(gain_num_short_square, gain_den, FRAC_BITS);
                        tmp2 = MULL(gain_num_square, gain_den_short, FRAC_BITS);

                        // R'(T)^2 > max R'(T)^2
                        if (tmp > tmp2) {
                            gain_num = gain_num_short;
                            gain_den = gain_den_short;
                            gain_num_square = gain_num_short_square;
                            delayed_signal_offset = i;
                            best_delay_frac = k + 1;
                        }
                    }
                }

                /*
R'(T)^2
2 * --------- < 1
R(0)
*/
                L64_temp0 = (int64_t)gain_num_square << ((sh_gain_num << 1) + 1);
                L64_temp1 = ((int64_t)gain_den * ener) << (sh_gain_den + sh_ener);
                if (L64_temp0 < L64_temp1)
                    gain_num = 0;
            } // if(sh_gain_den >= 0)
        } // if(corr_int_num)
    } // if(ener)
    /* End of best delay searching code */

    if (!gain_num) {
        memcpy(residual_filt, residual + RES_PREV_DATA_SIZE, subframe_size * sizeof(int16_t));

        /* Long-term prediction gain is less than 3dB. Long-term postfilter is disabled. */
        return 0;
    }
    if (best_delay_frac) {
        /* Recompute delayed signal with an interpolation filter of length 129. */
        ff_acelp_interpolate(residual_filt,
                             &sig_scaled[RES_PREV_DATA_SIZE - best_delay_int + delayed_signal_offset],
                             ff_g729_interp_filt_long,
                             ANALYZED_FRAC_DELAYS + 1,
                             8 - best_delay_frac,
                             LONG_INT_FILT_LEN,
                             subframe_size + 1);
        /* Compute R'(k) correlation's numerator. */
        sum = dsp->scalarproduct_int16(residual_filt,
                                       sig_scaled + RES_PREV_DATA_SIZE,
                                       subframe_size);

        if (sum < 0) {
            gain_long_num = 0;
            sh_gain_long_num = 0;
        } else {
            tmp = FFMAX(av_log2(sum) - 14, 0);
            sum >>= tmp;
            gain_long_num = sum;
            sh_gain_long_num = tmp;
        }

        /* Compute R'(k) correlation's denominator. */
        sum = dsp->scalarproduct_int16(residual_filt, residual_filt, subframe_size);

        tmp = FFMAX(av_log2(sum) - 14, 0);
        sum >>= tmp;
        gain_long_den = sum;
        sh_gain_long_den = tmp;

        /* Select between original and delayed signal.
Delayed signal will be selected if it increases R'(k)
correlation. */
        L_temp0 = gain_num * gain_num;
        L_temp0 = MULL(L_temp0, gain_long_den, FRAC_BITS);

        L_temp1 = gain_long_num * gain_long_num;
        L_temp1 = MULL(L_temp1, gain_den, FRAC_BITS);

        tmp = ((sh_gain_long_num - sh_gain_num) << 1) - (sh_gain_long_den - sh_gain_den);
        if (tmp > 0)
            L_temp0 >>= tmp;
        else
            L_temp1 >>= -tmp;

        /* Check if longer filter increases the values of R'(k). */
        if (L_temp1 > L_temp0) {
            /* Select long filter. */
            selected_signal = residual_filt;
            gain_num = gain_long_num;
            gain_den = gain_long_den;
            sh_gain_num = sh_gain_long_num;
            sh_gain_den = sh_gain_long_den;
        } else
            /* Select short filter. */
            selected_signal = &delayed_signal[best_delay_frac-1][delayed_signal_offset];

        /* Rescale selected signal to original value. */
        if (shift > 0)
            for (i = 0; i < subframe_size; i++)
                selected_signal[i] <<= shift;
        else
            for (i = 0; i < subframe_size; i++)
                selected_signal[i] >>= -shift;

        /* necessary to avoid compiler warning */
        selected_signal_const = selected_signal;
    } // if(best_delay_frac)
    else
        selected_signal_const = residual + RES_PREV_DATA_SIZE - (best_delay_int + 1 - delayed_signal_offset);
#ifdef G729_BITEXACT
    tmp = sh_gain_num - sh_gain_den;
    if (tmp > 0)
        gain_den >>= tmp;
    else
        gain_num >>= -tmp;

    if (gain_num > gain_den)
        lt_filt_factor_a = MIN_LT_FILT_FACTOR_A;
    else {
        gain_num >>= 2;
        gain_den >>= 1;
        lt_filt_factor_a = (gain_den << 15) / (gain_den + gain_num);
    }
#else
    L64_temp0 = ((int64_t)gain_num) << (sh_gain_num - 1);
    L64_temp1 = ((int64_t)gain_den) << sh_gain_den;
    lt_filt_factor_a = FFMAX((L64_temp1 << 15) / (L64_temp1 + L64_temp0), MIN_LT_FILT_FACTOR_A);
#endif

    /* Filter through selected filter. */
    lt_filt_factor_b = 32767 - lt_filt_factor_a + 1;

    ff_acelp_weighted_vector_sum(residual_filt, residual + RES_PREV_DATA_SIZE,
                                 selected_signal_const,
                                 lt_filt_factor_a, lt_filt_factor_b,
                                 1<<14, 15, subframe_size);

    // Long-term prediction gain is larger than 3dB.
    return 1;
}

/**
* \brief Calculate reflection coefficient for tilt compensation filter (4.2.3).
* \param dsp initialized DSP context
* \param lp_gn (3.12) coefficients of A(z/FORMANT_PP_FACTOR_NUM) filter
* \param lp_gd (3.12) coefficients of A(z/FORMANT_PP_FACTOR_DEN) filter
* \param speech speech to update
* \param subframe_size size of subframe
*
* \return (3.12) reflection coefficient
*
* \remark The routine also calculates the gain term for the short-term
* filter (gf) and multiplies the speech data by 1/gf.
*
* \note All members of lp_gn, except 10-19 must be equal to zero.
*/
static int16_t get_tilt_comp(DSPContext *dsp, int16_t *lp_gn,
                             const int16_t *lp_gd, int16_t* speech,
                             int subframe_size)
{
    int rh1,rh0; // (3.12)
    int temp;
    int i;
    int gain_term;

    lp_gn[10] = 4096; //1.0 in (3.12)

    /* Apply 1/A(z/FORMANT_PP_FACTOR_DEN) filter to hf. */
    ff_celp_lp_synthesis_filter(lp_gn + 11, lp_gd + 1, lp_gn + 11, 22, 10, 0, 0, 0x800);
    /* Now lp_gn (starting with 10) contains impulse response
of A(z/FORMANT_PP_FACTOR_NUM)/A(z/FORMANT_PP_FACTOR_DEN) filter. */

    rh0 = dsp->scalarproduct_int16(lp_gn + 10, lp_gn + 10, 20);
    rh1 = dsp->scalarproduct_int16(lp_gn + 10, lp_gn + 11, 20);

    /* downscale to avoid overflow */
    temp = av_log2(rh0) - 14;
    if (temp > 0) {
        rh0 >>= temp;
        rh1 >>= temp;
    }

    if (FFABS(rh1) > rh0 || !rh0)
        return 0;

    gain_term = 0;
    for (i = 0; i < 20; i++)
        gain_term += FFABS(lp_gn[i + 10]);
    gain_term >>= 2; // (3.12) -> (5.10)

    if (gain_term > 0x400) { // 1.0 in (5.10)
        temp = 0x2000000 / gain_term; // 1.0/gain_term in (0.15)
        for (i = 0; i < subframe_size; i++)
            speech[i] = (speech[i] * temp + 0x4000) >> 15;
    }

    return -(rh1 << 15) / rh0;
}

/**
* \brief Apply tilt compensation filter (4.2.3).
* \param res_pst [in/out] residual signal (partially filtered)
* \param k1 (3.12) reflection coefficient
* \param subframe_size size of subframe
* \param ht_prev_data previous data for 4.2.3, equation 86
*
* \return new value for ht_prev_data
*/
static int16_t apply_tilt_comp(int16_t* out, int16_t* res_pst, int refl_coeff,
                               int subframe_size, int16_t ht_prev_data)
{
    int tmp, tmp2;
    int i;
    int gt, ga;
    int fact, sh_fact;

    if (refl_coeff > 0) {
        gt = (refl_coeff * G729_TILT_FACTOR_PLUS + 0x4000) >> 15;
        fact = 0x4000; // 0.5 in (0.15)
        sh_fact = 15;
    } else {
        gt = (refl_coeff * G729_TILT_FACTOR_MINUS + 0x4000) >> 15;
        fact = 0x800; // 0.5 in (3.12)
        sh_fact = 12;
    }
    ga = (fact << 15) / av_clip_int16(32768 - FFABS(gt));
    gt >>= 1;

    /* Apply tilt compensation filter to signal. */
    tmp = res_pst[subframe_size - 1];

    for (i = subframe_size - 1; i >= 1; i--) {
        tmp2 = (res_pst[i] << 15) + ((gt * res_pst[i-1]) << 1);
        tmp2 = (tmp2 + 0x4000) >> 15;

        tmp2 = (tmp2 * ga * 2 + fact) >> sh_fact;
        out[i] = tmp2;
    }
    tmp2 = (res_pst[0] << 15) + ((gt * ht_prev_data) << 1);
    tmp2 = (tmp2 + 0x4000) >> 15;
    tmp2 = (tmp2 * ga * 2 + fact) >> sh_fact;
    out[0] = tmp2;

    return tmp;
}

void ff_g729_postfilter(DSPContext *dsp, int16_t* ht_prev_data, int* voicing,
                     const int16_t *lp_filter_coeffs, int pitch_delay_int,
                     int16_t* residual, int16_t* res_filter_data,
                     int16_t* pos_filter_data, int16_t *speech, int subframe_size)
{
    int16_t residual_filt_buf[SUBFRAME_SIZE+11];
    int16_t lp_gn[33]; // (3.12)
    int16_t lp_gd[11]; // (3.12)
    int tilt_comp_coeff;
    int i;

    /* Zero-filling is necessary for tilt-compensation filter. */
    memset(lp_gn, 0, 33 * sizeof(int16_t));

    /* Calculate A(z/FORMANT_PP_FACTOR_NUM) filter coefficients. */
    for (i = 0; i < 10; i++)
        lp_gn[i + 11] = (lp_filter_coeffs[i + 1] * formant_pp_factor_num_pow[i] + 0x4000) >> 15;

    /* Calculate A(z/FORMANT_PP_FACTOR_DEN) filter coefficients. */
    for (i = 0; i < 10; i++)
        lp_gd[i + 1] = (lp_filter_coeffs[i + 1] * formant_pp_factor_den_pow[i] + 0x4000) >> 15;

    /* residual signal calculation (one-half of short-term postfilter) */
    memcpy(speech - 10, res_filter_data, 10 * sizeof(int16_t));
    residual_filter(residual + RES_PREV_DATA_SIZE, lp_gn + 11, speech, subframe_size);
    /* Save data to use it in the next subframe. */
    memcpy(res_filter_data, speech + subframe_size - 10, 10 * sizeof(int16_t));

    /* long-term filter. If long-term prediction gain is larger than 3dB (returned value is
nonzero) then declare current subframe as periodic. */
    *voicing = FFMAX(*voicing, long_term_filter(dsp, pitch_delay_int,
                                                residual, residual_filt_buf + 10,
                                                subframe_size));

    /* shift residual for using in next subframe */
    memmove(residual, residual + subframe_size, RES_PREV_DATA_SIZE * sizeof(int16_t));

    /* short-term filter tilt compensation */
    tilt_comp_coeff = get_tilt_comp(dsp, lp_gn, lp_gd, residual_filt_buf + 10, subframe_size);

    /* Apply second half of short-term postfilter: 1/A(z/FORMANT_PP_FACTOR_DEN) */
    ff_celp_lp_synthesis_filter(pos_filter_data + 10, lp_gd + 1,
                                residual_filt_buf + 10,
                                subframe_size, 10, 0, 0, 0x800);
    memcpy(pos_filter_data, pos_filter_data + subframe_size, 10 * sizeof(int16_t));

    *ht_prev_data = apply_tilt_comp(speech, pos_filter_data + 10, tilt_comp_coeff,
                                    subframe_size, *ht_prev_data);
}

/**
* \brief Adaptive gain control (4.2.4)
* \param gain_before gain of speech before applying postfilters
* \param gain_after gain of speech after applying postfilters
* \param speech [in/out] signal buffer
* \param subframe_size length of subframe
* \param gain_prev (3.12) previous value of gain coefficient
*
* \return (3.12) last value of gain coefficient
*/
int16_t ff_g729_adaptive_gain_control(int gain_before, int gain_after, int16_t *speech,
                                   int subframe_size, int16_t gain_prev)
{
    int gain; // (3.12)
    int n;
    int exp_before, exp_after;

    if(!gain_after && gain_before)
        return 0;

    if (gain_before) {

        exp_before = 14 - av_log2(gain_before);
        gain_before = bidir_sal(gain_before, exp_before);

        exp_after = 14 - av_log2(gain_after);
        gain_after = bidir_sal(gain_after, exp_after);

        if (gain_before < gain_after) {
            gain = (gain_before << 15) / gain_after;
            gain = bidir_sal(gain, exp_after - exp_before - 1);
        } else {
            gain = ((gain_before - gain_after) << 14) / gain_after + 0x4000;
            gain = bidir_sal(gain, exp_after - exp_before);
        }
        gain = (gain * G729_AGC_FAC1 + 0x4000) >> 15; // gain * (1-0.9875)
    } else
        gain = 0;

    for (n = 0; n < subframe_size; n++) {
        // gain_prev = gain + 0.9875 * gain_prev
        gain_prev = (G729_AGC_FACTOR * gain_prev + 0x4000) >> 15;
        gain_prev = av_clip_int16(gain + gain_prev);
        speech[n] = av_clip_int16((speech[n] * gain_prev + 0x2000) >> 14);
    }
    return gain_prev;
}
Something went wrong with that request. Please try again.