FFmpeg/libavcodec/ra288.c

/*
 * RealAudio 2.0 (28.8K)
 * Copyright (c) 2003 the ffmpeg project
 *
 * 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 "avcodec.h"
#define ALT_BITSTREAM_READER_LE
#include "bitstream.h"
#include "ra288.h"

typedef struct {
    float sp_lpc[36];   ///< LPC coefficients for speech data (spec: A)
    float gain_lpc[10]; ///< LPC coefficients for gain (spec: GB)
    int   phase;

    float sp_hist[111]; ///< Speech data history (spec: SB)

    /** Speech part of the gain autocorrelation (spec: REXP) */
    float sp_rec[37];

    float gain_hist[38];   ///< Log-gain history (spec: SBLG)

    /** Recursive part of the gain autocorrelation (spec: REXPLG) */
    float gain_rec[11];

    float sb[41];
    float lhist[10];
} RA288Context;

static inline float scalar_product_float(const float * v1, const float * v2,
                                         int size)
{
    float res = 0.;

    while (size--)
        res += *v1++ * *v2++;

    return res;
}

static void colmult(float *tgt, const float *m1, const float *m2, int n)
{
    while (n--)
        *tgt++ = *m1++ * *m2++;
}

static void decode(RA288Context *ractx, float gain, int cb_coef)
{
    int x, y;
    double sumsum;
    float sum, buffer[5];

    memmove(ractx->sb + 5, ractx->sb, 36 * sizeof(*ractx->sb));

    for (x=4; x >= 0; x--)
        ractx->sb[x] = -scalar_product_float(ractx->sb + x + 1,
                                             ractx->sp_lpc, 36);

    /* block 46 of G.728 spec */
    sum = 32. - scalar_product_float(ractx->gain_lpc, ractx->lhist, 10);

    /* block 47 of G.728 spec */
    sum = av_clipf(sum, 0, 60);

    /* block 48 of G.728 spec */
    sumsum = exp(sum * 0.1151292546497) * gain;    /* pow(10.0,sum/20)*f */

    for (x=0; x < 5; x++)
        buffer[x] = codetable[cb_coef][x] * sumsum;

    sum = scalar_product_float(buffer, buffer, 5) / 5;

    sum = FFMAX(sum, 1);

    /* shift and store */
    memmove(ractx->lhist, ractx->lhist - 1, 10 * sizeof(*ractx->lhist));

    *ractx->lhist = 10 * log10(sum) - 32;

    for (x=1; x < 5; x++)
        for (y=x-1; y >= 0; y--)
            buffer[x] -= ractx->sp_lpc[x-y-1] * buffer[y];

    /* output */
    for (x=0; x < 5; x++)
        ractx->sb[4-x] = av_clipf(ractx->sb[4-x] + buffer[x], -4095, 4095);
}

/**
 * Converts autocorrelation coefficients to LPC coefficients using the
 * Levinson-Durbin algorithm. See blocks 37 and 50 of the G.728 specification.
 *
 * @return 0 if success, -1 if fail
 */
static int eval_lpc_coeffs(const float *in, float *tgt, int n)
{
    int x, y;
    double f0, f1, f2;

    if (in[n] == 0)
        return -1;

    if ((f0 = *in) <= 0)
        return -1;

    in--; // To avoid a -1 subtraction in the inner loop

    for (x=1; x <= n; x++) {
        f1 = in[x+1];

        for (y=0; y < x - 1; y++)
            f1 += in[x-y]*tgt[y];

        tgt[x-1] = f2 = -f1/f0;
        for (y=0; y < x >> 1; y++) {
            float temp = tgt[y] + tgt[x-y-2]*f2;
            tgt[x-y-2] += tgt[y]*f2;
            tgt[y] = temp;
        }
        if ((f0 += f1*f2) < 0)
            return -1;
    }

    return 0;
}

static void prodsum(float *tgt, const float *src, int len, int n)
{
    for (; n >= 0; n--)
        tgt[n] = scalar_product_float(src, src - n, len);

}

/**
 * Hybrid window filtering. See blocks 36 and 49 of the G.728 specification.
 *
 * @note This function is slightly different from that described in the spec.
 *       It expects in[0] to be the newest sample and in[n-1] to be the oldest
 *       one stored. The spec has in the more ordinary way (in[0] the oldest
 *       and in[n-1] the newest).
 *
 * @param order   the order of the filter
 * @param n       the length of the input
 * @param non_rec the number of non-recursive samples
 * @param out     the filter output
 * @param in      pointer to the input of the filter
 * @param hist    pointer to the input history of the filter. It is updated by
 *                this function.
 * @param out     pointer to the non-recursive part of the output
 * @param out2    pointer to the recursive part of the output
 * @param window  pointer to the windowing function table
 */
static void do_hybrid_window(int order, int n, int non_rec, const float *in,
                             float *out, float *hist, float *out2,
                             const float *window)
{
    unsigned int x;
    float buffer1[order + 1];
    float buffer2[order + 1];
    float work[order + n + non_rec];

    /* update history */
    memmove(hist, hist + n, (order + non_rec)*sizeof(*hist));

    for (x=0; x < n; x++)
        hist[order + non_rec + x] = in[n-x-1];

    colmult(work, window, hist, order + n + non_rec);

    prodsum(buffer1, work + order    , n      , order);
    prodsum(buffer2, work + order + n, non_rec, order);

    for (x=0; x <= order; x++) {
        out2[x] = out2[x] * 0.5625 + buffer1[x];
        out [x] = out2[x]          + buffer2[x];
    }

    /* Multiply by the white noise correcting factor (WNCF) */
    *out *= 257./256.;
}

/**
 * Backward synthesis filter. Find the LPC coefficients from past speech data.
 */
static void backward_filter(RA288Context *ractx)
{
    float temp1[37]; // RTMP in the spec
    float temp2[11]; // GPTPMP in the spec

    do_hybrid_window(36, 40, 35, ractx->sb, temp1, ractx->sp_hist,
                     ractx->sp_rec, syn_window);

    if (!eval_lpc_coeffs(temp1, ractx->sp_lpc, 36))
        colmult(ractx->sp_lpc, ractx->sp_lpc, syn_bw_tab, 36);

    do_hybrid_window(10, 8, 20, ractx->lhist, temp2, ractx->gain_hist,
                     ractx->gain_rec, gain_window);

    if (!eval_lpc_coeffs(temp2, ractx->gain_lpc, 10))
        colmult(ractx->gain_lpc, ractx->gain_lpc, gain_bw_tab, 10);
}

static int ra288_decode_frame(AVCodecContext * avctx, void *data,
                              int *data_size, const uint8_t * buf,
                              int buf_size)
{
    int16_t *out = data;
    int x, y;
    RA288Context *ractx = avctx->priv_data;
    GetBitContext gb;

    if (buf_size < avctx->block_align) {
        av_log(avctx, AV_LOG_ERROR,
               "Error! Input buffer is too small [%d<%d]\n",
               buf_size, avctx->block_align);
        return 0;
    }

    init_get_bits(&gb, buf, avctx->block_align * 8);

    for (x=0; x < 32; x++) {
        float gain = amptable[get_bits(&gb, 3)];
        int cb_coef = get_bits(&gb, 6 + (x&1));
        ractx->phase = (x + 4) & 7;
        decode(ractx, gain, cb_coef);

        for (y=0; y < 5; y++)
            *(out++) = 8 * ractx->sb[4 - y];

        if (ractx->phase == 7)
            backward_filter(ractx);
    }

    *data_size = (char *)out - (char *)data;
    return avctx->block_align;
}

AVCodec ra_288_decoder =
{
    "real_288",
    CODEC_TYPE_AUDIO,
    CODEC_ID_RA_288,
    sizeof(RA288Context),
    NULL,
    NULL,
    NULL,
    ra288_decode_frame,
    .long_name = NULL_IF_CONFIG_SMALL("RealAudio 2.0 (28.8K)"),
};