Skip to content
This repository

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Fetching contributors…

Cannot retrieve contributors at this time

file 500 lines (420 sloc) 15.422 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
/*
* G.726 ADPCM audio codec
* Copyright (c) 2004 Roman Shaposhnik
*
* This is a very straightforward rendition of the G.726
* Section 4 "Computational Details".
*
* 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 <limits.h>
#include "libavutil/avassert.h"
#include "libavutil/opt.h"
#include "avcodec.h"
#include "internal.h"
#include "get_bits.h"
#include "put_bits.h"

/**
* G.726 11bit float.
* G.726 Standard uses rather odd 11bit floating point arithmentic for
* numerous occasions. It's a mystery to me why they did it this way
* instead of simply using 32bit integer arithmetic.
*/
typedef struct Float11 {
    uint8_t sign; /**< 1bit sign */
    uint8_t exp; /**< 4bit exponent */
    uint8_t mant; /**< 6bit mantissa */
} Float11;

static inline Float11* i2f(int i, Float11* f)
{
    f->sign = (i < 0);
    if (f->sign)
        i = -i;
    f->exp = av_log2_16bit(i) + !!i;
    f->mant = i? (i<<6) >> f->exp : 1<<5;
    return f;
}

static inline int16_t mult(Float11* f1, Float11* f2)
{
        int res, exp;

        exp = f1->exp + f2->exp;
        res = (((f1->mant * f2->mant) + 0x30) >> 4);
        res = exp > 19 ? res << (exp - 19) : res >> (19 - exp);
        return (f1->sign ^ f2->sign) ? -res : res;
}

static inline int sgn(int value)
{
    return (value < 0) ? -1 : 1;
}

typedef struct G726Tables {
    const int* quant; /**< quantization table */
    const int16_t* iquant; /**< inverse quantization table */
    const int16_t* W; /**< special table #1 ;-) */
    const uint8_t* F; /**< special table #2 */
} G726Tables;

typedef struct G726Context {
    AVClass *class;
    AVFrame frame;
    G726Tables tbls; /**< static tables needed for computation */

    Float11 sr[2]; /**< prev. reconstructed samples */
    Float11 dq[6]; /**< prev. difference */
    int a[2]; /**< second order predictor coeffs */
    int b[6]; /**< sixth order predictor coeffs */
    int pk[2]; /**< signs of prev. 2 sez + dq */

    int ap; /**< scale factor control */
    int yu; /**< fast scale factor */
    int yl; /**< slow scale factor */
    int dms; /**< short average magnitude of F[i] */
    int dml; /**< long average magnitude of F[i] */
    int td; /**< tone detect */

    int se; /**< estimated signal for the next iteration */
    int sez; /**< estimated second order prediction */
    int y; /**< quantizer scaling factor for the next iteration */
    int code_size;
} G726Context;

static const int quant_tbl16[] = /**< 16kbit/s 2bits per sample */
           { 260, INT_MAX };
static const int16_t iquant_tbl16[] =
           { 116, 365, 365, 116 };
static const int16_t W_tbl16[] =
           { -22, 439, 439, -22 };
static const uint8_t F_tbl16[] =
           { 0, 7, 7, 0 };

static const int quant_tbl24[] = /**< 24kbit/s 3bits per sample */
           { 7, 217, 330, INT_MAX };
static const int16_t iquant_tbl24[] =
           { INT16_MIN, 135, 273, 373, 373, 273, 135, INT16_MIN };
static const int16_t W_tbl24[] =
           { -4, 30, 137, 582, 582, 137, 30, -4 };
static const uint8_t F_tbl24[] =
           { 0, 1, 2, 7, 7, 2, 1, 0 };

static const int quant_tbl32[] = /**< 32kbit/s 4bits per sample */
           { -125, 79, 177, 245, 299, 348, 399, INT_MAX };
static const int16_t iquant_tbl32[] =
         { INT16_MIN, 4, 135, 213, 273, 323, 373, 425,
                 425, 373, 323, 273, 213, 135, 4, INT16_MIN };
static const int16_t W_tbl32[] =
           { -12, 18, 41, 64, 112, 198, 355, 1122,
            1122, 355, 198, 112, 64, 41, 18, -12};
static const uint8_t F_tbl32[] =
           { 0, 0, 0, 1, 1, 1, 3, 7, 7, 3, 1, 1, 1, 0, 0, 0 };

static const int quant_tbl40[] = /**< 40kbit/s 5bits per sample */
           { -122, -16, 67, 138, 197, 249, 297, 338,
              377, 412, 444, 474, 501, 527, 552, INT_MAX };
static const int16_t iquant_tbl40[] =
         { INT16_MIN, -66, 28, 104, 169, 224, 274, 318,
                 358, 395, 429, 459, 488, 514, 539, 566,
                 566, 539, 514, 488, 459, 429, 395, 358,
                 318, 274, 224, 169, 104, 28, -66, INT16_MIN };
static const int16_t W_tbl40[] =
           { 14, 14, 24, 39, 40, 41, 58, 100,
              141, 179, 219, 280, 358, 440, 529, 696,
              696, 529, 440, 358, 280, 219, 179, 141,
              100, 58, 41, 40, 39, 24, 14, 14 };
static const uint8_t F_tbl40[] =
           { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 3, 4, 5, 6, 6,
             6, 6, 5, 4, 3, 2, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 };

static const G726Tables G726Tables_pool[] =
           {{ quant_tbl16, iquant_tbl16, W_tbl16, F_tbl16 },
            { quant_tbl24, iquant_tbl24, W_tbl24, F_tbl24 },
            { quant_tbl32, iquant_tbl32, W_tbl32, F_tbl32 },
            { quant_tbl40, iquant_tbl40, W_tbl40, F_tbl40 }};


/**
* Para 4.2.2 page 18: Adaptive quantizer.
*/
static inline uint8_t quant(G726Context* c, int d)
{
    int sign, exp, i, dln;

    sign = i = 0;
    if (d < 0) {
        sign = 1;
        d = -d;
    }
    exp = av_log2_16bit(d);
    dln = ((exp<<7) + (((d<<7)>>exp)&0x7f)) - (c->y>>2);

    while (c->tbls.quant[i] < INT_MAX && c->tbls.quant[i] < dln)
        ++i;

    if (sign)
        i = ~i;
    if (c->code_size != 2 && i == 0) /* I'm not sure this is a good idea */
        i = 0xff;

    return i;
}

/**
* Para 4.2.3 page 22: Inverse adaptive quantizer.
*/
static inline int16_t inverse_quant(G726Context* c, int i)
{
    int dql, dex, dqt;

    dql = c->tbls.iquant[i] + (c->y >> 2);
    dex = (dql>>7) & 0xf; /* 4bit exponent */
    dqt = (1<<7) + (dql & 0x7f); /* log2 -> linear */
    return (dql < 0) ? 0 : ((dqt<<dex) >> 7);
}

static int16_t g726_decode(G726Context* c, int I)
{
    int dq, re_signal, pk0, fa1, i, tr, ylint, ylfrac, thr2, al, dq0;
    Float11 f;
    int I_sig= I >> (c->code_size - 1);

    dq = inverse_quant(c, I);

    /* Transition detect */
    ylint = (c->yl >> 15);
    ylfrac = (c->yl >> 10) & 0x1f;
    thr2 = (ylint > 9) ? 0x1f << 10 : (0x20 + ylfrac) << ylint;
    tr= (c->td == 1 && dq > ((3*thr2)>>2));

    if (I_sig) /* get the sign */
        dq = -dq;
    re_signal = c->se + dq;

    /* Update second order predictor coefficient A2 and A1 */
    pk0 = (c->sez + dq) ? sgn(c->sez + dq) : 0;
    dq0 = dq ? sgn(dq) : 0;
    if (tr) {
        c->a[0] = 0;
        c->a[1] = 0;
        for (i=0; i<6; i++)
            c->b[i] = 0;
    } else {
        /* This is a bit crazy, but it really is +255 not +256 */
        fa1 = av_clip((-c->a[0]*c->pk[0]*pk0)>>5, -256, 255);

        c->a[1] += 128*pk0*c->pk[1] + fa1 - (c->a[1]>>7);
        c->a[1] = av_clip(c->a[1], -12288, 12288);
        c->a[0] += 64*3*pk0*c->pk[0] - (c->a[0] >> 8);
        c->a[0] = av_clip(c->a[0], -(15360 - c->a[1]), 15360 - c->a[1]);

        for (i=0; i<6; i++)
            c->b[i] += 128*dq0*sgn(-c->dq[i].sign) - (c->b[i]>>8);
    }

    /* Update Dq and Sr and Pk */
    c->pk[1] = c->pk[0];
    c->pk[0] = pk0 ? pk0 : 1;
    c->sr[1] = c->sr[0];
    i2f(re_signal, &c->sr[0]);
    for (i=5; i>0; i--)
        c->dq[i] = c->dq[i-1];
    i2f(dq, &c->dq[0]);
    c->dq[0].sign = I_sig; /* Isn't it crazy ?!?! */

    c->td = c->a[1] < -11776;

    /* Update Ap */
    c->dms += (c->tbls.F[I]<<4) + ((- c->dms) >> 5);
    c->dml += (c->tbls.F[I]<<4) + ((- c->dml) >> 7);
    if (tr)
        c->ap = 256;
    else {
        c->ap += (-c->ap) >> 4;
        if (c->y <= 1535 || c->td || abs((c->dms << 2) - c->dml) >= (c->dml >> 3))
            c->ap += 0x20;
    }

    /* Update Yu and Yl */
    c->yu = av_clip(c->y + c->tbls.W[I] + ((-c->y)>>5), 544, 5120);
    c->yl += c->yu + ((-c->yl)>>6);

    /* Next iteration for Y */
    al = (c->ap >= 256) ? 1<<6 : c->ap >> 2;
    c->y = (c->yl + (c->yu - (c->yl>>6))*al) >> 6;

    /* Next iteration for SE and SEZ */
    c->se = 0;
    for (i=0; i<6; i++)
        c->se += mult(i2f(c->b[i] >> 2, &f), &c->dq[i]);
    c->sez = c->se >> 1;
    for (i=0; i<2; i++)
        c->se += mult(i2f(c->a[i] >> 2, &f), &c->sr[i]);
    c->se >>= 1;

    return av_clip(re_signal << 2, -0xffff, 0xffff);
}

static av_cold int g726_reset(G726Context *c)
{
    int i;

    c->tbls = G726Tables_pool[c->code_size - 2];
    for (i=0; i<2; i++) {
        c->sr[i].mant = 1<<5;
        c->pk[i] = 1;
    }
    for (i=0; i<6; i++) {
        c->dq[i].mant = 1<<5;
    }
    c->yu = 544;
    c->yl = 34816;

    c->y = 544;

    return 0;
}

#if CONFIG_ADPCM_G726_ENCODER
static int16_t g726_encode(G726Context* c, int16_t sig)
{
    uint8_t i;

    i = quant(c, sig/4 - c->se) & ((1<<c->code_size) - 1);
    g726_decode(c, i);
    return i;
}

/* Interfacing to the libavcodec */

static av_cold int g726_encode_init(AVCodecContext *avctx)
{
    G726Context* c = avctx->priv_data;

    if (avctx->strict_std_compliance > FF_COMPLIANCE_UNOFFICIAL &&
        avctx->sample_rate != 8000) {
        av_log(avctx, AV_LOG_ERROR, "Sample rates other than 8kHz are not "
               "allowed when the compliance level is higher than unofficial. "
               "Resample or reduce the compliance level.\n");
        return AVERROR(EINVAL);
    }
    av_assert0(avctx->sample_rate > 0);

    if(avctx->channels != 1){
        av_log(avctx, AV_LOG_ERROR, "Only mono is supported\n");
        return AVERROR(EINVAL);
    }

    if (avctx->bit_rate)
        c->code_size = (avctx->bit_rate + avctx->sample_rate/2) / avctx->sample_rate;

    c->code_size = av_clip(c->code_size, 2, 5);
    avctx->bit_rate = c->code_size * avctx->sample_rate;
    avctx->bits_per_coded_sample = c->code_size;

    g726_reset(c);

#if FF_API_OLD_ENCODE_AUDIO
    avctx->coded_frame = avcodec_alloc_frame();
    if (!avctx->coded_frame)
        return AVERROR(ENOMEM);
    avctx->coded_frame->key_frame = 1;
#endif

    /* select a frame size that will end on a byte boundary and have a size of
approximately 1024 bytes */
    avctx->frame_size = ((int[]){ 4096, 2736, 2048, 1640 })[c->code_size - 2];

    return 0;
}

#if FF_API_OLD_ENCODE_AUDIO
static av_cold int g726_encode_close(AVCodecContext *avctx)
{
    av_freep(&avctx->coded_frame);
    return 0;
}
#endif

static int g726_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
                             const AVFrame *frame, int *got_packet_ptr)
{
    G726Context *c = avctx->priv_data;
    const int16_t *samples = (const int16_t *)frame->data[0];
    PutBitContext pb;
    int i, ret, out_size;

    out_size = (frame->nb_samples * c->code_size + 7) / 8;
    if ((ret = ff_alloc_packet2(avctx, avpkt, out_size)))
        return ret;
    init_put_bits(&pb, avpkt->data, avpkt->size);

    for (i = 0; i < frame->nb_samples; i++)
        put_bits(&pb, c->code_size, g726_encode(c, *samples++));

    flush_put_bits(&pb);

    avpkt->size = out_size;
    *got_packet_ptr = 1;
    return 0;
}

#define OFFSET(x) offsetof(G726Context, x)
#define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
static const AVOption options[] = {
    { "code_size", "Bits per code", OFFSET(code_size), AV_OPT_TYPE_INT, { .i64 = 4 }, 2, 5, AE },
    { NULL },
};

static const AVClass class = {
    .class_name = "g726",
    .item_name = av_default_item_name,
    .option = options,
    .version = LIBAVUTIL_VERSION_INT,
};

static const AVCodecDefault defaults[] = {
    { "b", "0" },
    { NULL },
};

AVCodec ff_adpcm_g726_encoder = {
    .name = "g726",
    .type = AVMEDIA_TYPE_AUDIO,
    .id = AV_CODEC_ID_ADPCM_G726,
    .priv_data_size = sizeof(G726Context),
    .init = g726_encode_init,
    .encode2 = g726_encode_frame,
#if FF_API_OLD_ENCODE_AUDIO
    .close = g726_encode_close,
#endif
    .capabilities = CODEC_CAP_SMALL_LAST_FRAME,
    .sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_S16,
                                                     AV_SAMPLE_FMT_NONE },
    .long_name = NULL_IF_CONFIG_SMALL("G.726 ADPCM"),
    .priv_class = &class,
    .defaults = defaults,
};
#endif

#if CONFIG_ADPCM_G726_DECODER
static av_cold int g726_decode_init(AVCodecContext *avctx)
{
    G726Context* c = avctx->priv_data;

    if (avctx->strict_std_compliance >= FF_COMPLIANCE_STRICT &&
        avctx->sample_rate != 8000) {
        av_log(avctx, AV_LOG_ERROR, "Only 8kHz sample rate is allowed when "
               "the compliance level is strict. Reduce the compliance level "
               "if you wish to decode the stream anyway.\n");
        return AVERROR(EINVAL);
    }

    if(avctx->channels != 1){
        av_log(avctx, AV_LOG_ERROR, "Only mono is supported\n");
        return AVERROR(EINVAL);
    }

    c->code_size = avctx->bits_per_coded_sample;
    if (c->code_size < 2 || c->code_size > 5) {
        av_log(avctx, AV_LOG_ERROR, "Invalid number of bits %d\n", c->code_size);
        return AVERROR(EINVAL);
    }
    g726_reset(c);

    avctx->sample_fmt = AV_SAMPLE_FMT_S16;

    avcodec_get_frame_defaults(&c->frame);
    avctx->coded_frame = &c->frame;

    return 0;
}

static int g726_decode_frame(AVCodecContext *avctx, void *data,
                             int *got_frame_ptr, AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size = avpkt->size;
    G726Context *c = avctx->priv_data;
    int16_t *samples;
    GetBitContext gb;
    int out_samples, ret;

    out_samples = buf_size * 8 / c->code_size;

    /* get output buffer */
    c->frame.nb_samples = out_samples;
    if ((ret = avctx->get_buffer(avctx, &c->frame)) < 0) {
        av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
        return ret;
    }
    samples = (int16_t *)c->frame.data[0];

    init_get_bits(&gb, buf, buf_size * 8);

    while (out_samples--)
        *samples++ = g726_decode(c, get_bits(&gb, c->code_size));

    if (get_bits_left(&gb) > 0)
        av_log(avctx, AV_LOG_ERROR, "Frame invalidly split, missing parser?\n");

    *got_frame_ptr = 1;
    *(AVFrame *)data = c->frame;

    return buf_size;
}

static void g726_decode_flush(AVCodecContext *avctx)
{
    G726Context *c = avctx->priv_data;
    g726_reset(c);
}

AVCodec ff_adpcm_g726_decoder = {
    .name = "g726",
    .type = AVMEDIA_TYPE_AUDIO,
    .id = AV_CODEC_ID_ADPCM_G726,
    .priv_data_size = sizeof(G726Context),
    .init = g726_decode_init,
    .decode = g726_decode_frame,
    .flush = g726_decode_flush,
    .capabilities = CODEC_CAP_DR1,
    .long_name = NULL_IF_CONFIG_SMALL("G.726 ADPCM"),
};
#endif
Something went wrong with that request. Please try again.