[FFmpeg-devel] [PATCH] G722 decoder
Kenan Gillet
kenan.gillet
Sat Mar 21 08:43:35 CET 2009
Hi,
Here is an implementation of a bitexact G.722 decoder.
It is based upon the patch G722_decoder_by_Chas_Williams at [1]
and has been tested against the reference implementation found in ITU-G.711 .
A sample can be found at /MPlayer/incoming/g722decoder
In order to play the file, we need to be able force the bitrate from
the comment line (patch #1)
and to use a specific demuxer (patch #2)
So far I have only found 64kbps / 16Khz raw samples, but i tested the
other bitrate (48kbps, 56 kbps)
against the reference implementation
you can test the uploaded sample with the command:
ffmpeg -ac 1 -ab 64000 -ar 16000 -i conf-adminmenu-162.g722 output.wav
Thanks,
Kenan
[1] http://wiki.multimedia.cx/index.php?title=Interesting_Patches#G722_decoder_by_Chas_Williams:
-------------- next part --------------
Index: ffmpeg.c
===================================================================
--- ffmpeg.c (revision 18096)
+++ ffmpeg.c (working copy)
@@ -150,6 +150,7 @@
static int intra_only = 0;
static int audio_sample_rate = 44100;
+static int audio_bit_rate;
static int64_t channel_layout = 0;
#define QSCALE_NONE -99999
static float audio_qscale = QSCALE_NONE;
@@ -2345,12 +2346,14 @@
static int opt_bitrate(const char *opt, const char *arg)
{
int codec_type = opt[0]=='a' ? CODEC_TYPE_AUDIO : CODEC_TYPE_VIDEO;
+ int bit_rate;
opt_default(opt, arg);
- if (av_get_int(avcodec_opts[codec_type], "b", NULL) < 1000)
+ if ((bit_rate = av_get_int(avcodec_opts[codec_type], "b", NULL)) < 1000)
fprintf(stderr, "WARNING: The bitrate parameter is set too low. It takes bits/s as argument, not kbits/s\n");
-
+ if (codec_type == CODEC_TYPE_AUDIO)
+ audio_bit_rate = bit_rate;
return 0;
}
@@ -2819,6 +2822,7 @@
memset(ap, 0, sizeof(*ap));
ap->prealloced_context = 1;
ap->sample_rate = audio_sample_rate;
+ ap->bit_rate = audio_bit_rate;
ap->channels = audio_channels;
ap->time_base.den = frame_rate.num;
ap->time_base.num = frame_rate.den;
@@ -2892,6 +2896,7 @@
channel_layout = enc->channel_layout;
audio_channels = enc->channels;
audio_sample_rate = enc->sample_rate;
+ audio_bit_rate = enc->bit_rate;
audio_sample_fmt = enc->sample_fmt;
input_codecs[nb_icodecs++] = avcodec_find_decoder_by_name(audio_codec_name);
if(audio_disable)
@@ -3211,6 +3216,7 @@
}
nb_ocodecs++;
audio_enc->sample_rate = audio_sample_rate;
+ audio_enc->bit_rate = audio_bit_rate;
audio_enc->time_base= (AVRational){1, audio_sample_rate};
if (audio_language) {
av_metadata_set(&st->metadata, "language", audio_language);
Index: libavformat/avformat.h
===================================================================
--- libavformat/avformat.h (revision 18096)
+++ libavformat/avformat.h (working copy)
@@ -261,6 +261,7 @@
enum CodecID video_codec_id;
enum CodecID audio_codec_id;
#endif
+ int bit_rate;
} AVFormatParameters;
//! Demuxer will use url_fopen, no opened file should be provided by the caller.
Index: libavformat/raw.c
===================================================================
--- libavformat/raw.c (revision 18096)
+++ libavformat/raw.c (working copy)
@@ -82,6 +82,7 @@
switch(st->codec->codec_type) {
case CODEC_TYPE_AUDIO:
st->codec->sample_rate = ap->sample_rate;
+ st->codec->bit_rate = ap->bit_rate;
if(ap->channels) st->codec->channels = ap->channels;
else st->codec->channels = 1;
av_set_pts_info(st, 64, 1, st->codec->sample_rate);
-------------- next part --------------
Index: Changelog
===================================================================
--- Changelog (revision 18096)
+++ Changelog (working copy)
@@ -6,6 +6,7 @@
- VQF demuxer
- Alpha channel scaler
- PCX encoder
+- G.722 ADPCM audio decoder
Index: doc/general.texi
===================================================================
--- doc/general.texi (revision 18096)
+++ doc/general.texi (working copy)
@@ -482,6 +482,7 @@
@item ADPCM Electronic Arts R2 @tab @tab X
@item ADPCM Electronic Arts R3 @tab @tab X
@item ADPCM Electronic Arts XAS @tab @tab X
+ at item ADPCM G.722 @tab @tab X
@item ADPCM G.726 @tab X @tab X
@item ADPCM IMA AMV @tab @tab X
@tab Used in AMV files
Index: libavcodec/Makefile
===================================================================
--- libavcodec/Makefile (revision 18096)
+++ libavcodec/Makefile (working copy)
@@ -318,6 +318,7 @@
OBJS-$(CONFIG_ADPCM_EA_R2_DECODER) += adpcm.o
OBJS-$(CONFIG_ADPCM_EA_R3_DECODER) += adpcm.o
OBJS-$(CONFIG_ADPCM_EA_XAS_DECODER) += adpcm.o
+OBJS-$(CONFIG_ADPCM_G722_DECODER) += g722dec.o
OBJS-$(CONFIG_ADPCM_G726_DECODER) += g726.o
OBJS-$(CONFIG_ADPCM_G726_ENCODER) += g726.o
OBJS-$(CONFIG_ADPCM_IMA_AMV_DECODER) += adpcm.o
Index: libavcodec/allcodecs.c
===================================================================
--- libavcodec/allcodecs.c (revision 18096)
+++ libavcodec/allcodecs.c (working copy)
@@ -268,6 +268,7 @@
REGISTER_DECODER (ADPCM_EA_R3, adpcm_ea_r3);
REGISTER_DECODER (ADPCM_EA_XAS, adpcm_ea_xas);
REGISTER_ENCDEC (ADPCM_G726, adpcm_g726);
+ REGISTER_DECODER (ADPCM_G722, adpcm_g722);
REGISTER_DECODER (ADPCM_IMA_AMV, adpcm_ima_amv);
REGISTER_DECODER (ADPCM_IMA_DK3, adpcm_ima_dk3);
REGISTER_DECODER (ADPCM_IMA_DK4, adpcm_ima_dk4);
Index: libavcodec/avcodec.h
===================================================================
--- libavcodec/avcodec.h (revision 18096)
+++ libavcodec/avcodec.h (working copy)
@@ -247,6 +247,7 @@
CODEC_ID_ADPCM_EA_XAS,
CODEC_ID_ADPCM_EA_MAXIS_XA,
CODEC_ID_ADPCM_IMA_ISS,
+ CODEC_ID_ADPCM_G722,
/* AMR */
CODEC_ID_AMR_NB= 0x12000,
Index: libavcodec/g722dec.c
===================================================================
--- libavcodec/g722dec.c (revision 0)
+++ libavcodec/g722dec.c (revision 0)
@@ -0,0 +1,275 @@
+/*
+ * G.722 ADPCM audio decoder
+ *
+ * Copyright (c) 2005 Steve Underwood <steveu at coppice.org>
+ * Copyright (c) CMU 1993 Computer Science, Speech Group
+ * Chengxiang Lu and Alex Hauptmann
+ *
+ * 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
+ */
+
+
+/**
+ * @file libavcodec/g722.c
+ *
+ * G.722 ADPCM audio codec
+ *
+ * This G.722 decoder is a bit exact implementation of the ITU G.722 specification
+ * for all three specified bit rates - 64000bps, 56000bps and 48000bps.
+ * It passes the ITU tests.
+ *
+ * @note For the 56000bps and 48000bps bitrates, the respectively 7 and 6 bits
+ * codeword might be packed, so unpacking might be needed either
+ * internally or as a seprate parser.
+ */
+
+#include <stdint.h>
+#include "avcodec.h"
+#include "mathops.h"
+
+typedef struct {
+ int bits_per_sample; ///< 6 for 48000kbps, 7 for 56000kbps, or 8 for 64000kbps
+
+ int16_t prev_samples[24]; ///< memory of past 24 received (decoded) samples
+
+ /**
+ * The band[0] and band[1] correspond respectively to the lower band and higher band.
+ */
+ struct G722Band {
+ int16_t s_predictor; ///< predictor output value
+ int32_t s_zero; ///< zero section output signal
+ int8_t part_reconst_mem[2]; ///< partially reconstructed signal memory
+ int16_t qtzd_reconst_mem[2]; ///< quantized reconstructed signal
+ int16_t pole_mem[2]; ///< second-order pole section coefficient buffer
+ int16_t diff_mem[6]; ///< quantizer difference signal memory
+ int16_t zero_mem[6]; ///< Seventh-order zero section coefficient buffer
+ int16_t log_factor; ///< delayed logarithmic quantizer factor
+ int16_t scale_factor; ///< delayed quantizer scale factor
+ } band[2];
+} G722Context;
+
+
+static const int sign_lookup[2] = { -1, 1 };
+
+static const int16_t ilb[32] = {
+ 2048, 2093, 2139, 2186, 2233, 2282, 2332, 2383,
+ 2435, 2489, 2543, 2599, 2656, 2714, 2774, 2834,
+ 2896, 2960, 3025, 3091, 3158, 3228, 3298, 3371,
+ 3444, 3520, 3597, 3676, 3756, 3838, 3922, 4008
+};
+static const int16_t wh[2] = { 798, -214 };
+static const int16_t qm2[4] = { -7408, -1616, 7408, 1616 };
+/**
+ * qm3[index] == wl[rl42[index]]
+ */
+static const int16_t qm3[16] = {
+ -60, 3042, 1198, 538, 334, 172, 58, -30,
+ 3042, 1198, 538, 334, 172, 58, -30, -60
+};
+static const int16_t qm4[16] = {
+ 0, -20456, -12896, -8968, -6288, -4240, -2584, -1200,
+ 20456, 12896, 8968, 6288, 4240, 2584, 1200, 0
+};
+static const int16_t qm5[32] = {
+ -280, -280, -23352, -17560, -14120, -11664, -9752, -8184,
+ -6864, -5712, -4696, -3784, -2960, -2208, -1520, -880,
+ 23352, 17560, 14120, 11664, 9752, 8184, 6864, 5712,
+ 4696, 3784, 2960, 2208, 1520, 880, 280, -280
+};
+static const int16_t qm6[64] = {
+ -136, -136, -136, -136, -24808, -21904, -19008, -16704,
+ -14984, -13512, -12280, -11192, -10232, -9360, -8576, -7856,
+ -7192, -6576, -6000, -5456, -4944, -4464, -4008, -3576,
+ -3168, -2776, -2400, -2032, -1688, -1360, -1040, -728,
+ 24808, 21904, 19008, 16704, 14984, 13512, 12280, 11192,
+ 10232, 9360, 8576, 7856, 7192, 6576, 6000, 5456,
+ 4944, 4464, 4008, 3576, 3168, 2776, 2400, 2032,
+ 1688, 1360, 1040, 728, 432, 136, -432, -136
+};
+
+/**
+ * quadrature mirror filters (QMF) coefficients
+ *
+ * ITU-T G.722 Table 11
+ */
+static const int16_t qmf_coeffs[12] = {
+ 3, -11, 12, 32, -210, 951, 3876, -805, 362, -156, 53, -11,
+};
+
+
+/**
+ * adpative preditor
+ *
+ * @note On x86 using the MULL macro in a loop is slower than not using the macro.
+ */
+static void do_adaptive_prediction(struct G722Band *band, const int cur_diff)
+{
+ int sg[2], limit, i, cur_part_reconst;
+
+ band->qtzd_reconst_mem[1] = band->qtzd_reconst_mem[0];
+ band->qtzd_reconst_mem[0] = av_clip_int16((band->s_predictor + cur_diff) << 1);
+
+ cur_part_reconst = band->s_zero + cur_diff < 0;
+
+ sg[0] = sign_lookup[cur_part_reconst != band->part_reconst_mem[0]];
+ sg[1] = sign_lookup[cur_part_reconst == band->part_reconst_mem[1]];
+ band->part_reconst_mem[1] = band->part_reconst_mem[0];
+ band->part_reconst_mem[0] = cur_part_reconst;
+
+ band->pole_mem[1] = av_clip((sg[0] * av_clip(band->pole_mem[0], -8191, 8191) >> 5) +
+ (sg[1] << 7) + MULL(band->pole_mem[1], 127, 7), -12288, 12288);
+
+ limit = 15360 - band->pole_mem[1];
+ band->pole_mem[0] = av_clip(-192 * sg[0] + MULL(band->pole_mem[0], 255, 8), -limit, limit);
+
+
+ if(cur_diff) {
+ for (i = 0; i < 6; i++)
+ band->zero_mem[i] = ((band->zero_mem[i]*255) >> 8) +
+ (band->diff_mem[i] >> 15 == cur_diff >> 15 ? 128 : -128);
+ } else
+ for (i = 0; i < 6; i++)
+ band->zero_mem[i] = (band->zero_mem[i]*255) >> 8;
+
+ for (i = 5; i > 0; i--)
+ band->diff_mem[i] = band->diff_mem[i-1];
+ band->diff_mem[0] = cur_diff;
+
+ band->s_zero = 0;
+ for (i = 5; i >= 0; i--)
+ band->s_zero += (band->zero_mem[i]*av_clip_int16(band->diff_mem[i] << 1)) >> 15;
+ band->s_zero = av_clip_int16(band->s_zero);
+
+
+ band->s_predictor = av_clip_int16(band->s_zero +
+ MULL(band->pole_mem[0], band->qtzd_reconst_mem[0], 15) +
+ MULL(band->pole_mem[1], band->qtzd_reconst_mem[1], 15));
+}
+
+static int inline scale(const int log_factor, int shift) {
+ const int wd1 = ilb[(log_factor >> 6) & 31];
+ shift -= log_factor >> 11;
+ return shift < 0 ? wd1 << (2-shift) : (wd1 >> shift) << 2;
+}
+
+static int g722_decode_frame(AVCodecContext *avctx,
+ void *data, int *data_size,
+ const uint8_t *buf, const int buf_size)
+{
+ G722Context *c = avctx->priv_data;
+ int16_t *out_buf = data;
+ int j, out_len = 0;
+
+ for (j = 0; j < buf_size; j++) {
+ const int ihigh = (buf[j] >> (c->bits_per_sample-2)) & 0x03;
+ int rlow, rhigh, ilow;
+
+ // inverse adaptive quantizer
+ if (c->bits_per_sample == 8) {
+ ilow = buf[j] & 0x3F;
+ rlow = qm6[ilow];
+ ilow >>= 2;
+ } else if (c->bits_per_sample == 7) {
+ ilow = buf[j] & 0x1F;
+ rlow = qm5[ilow];
+ ilow >>= 1;
+ } else {
+ ilow = buf[j] & 0x0F;
+ rlow = qm4[ilow];
+ }
+ rlow = av_clip(MULL(c->band[0].scale_factor, rlow, 15) + c->band[0].s_predictor,
+ -16384, 16383);
+
+ do_adaptive_prediction(&c->band[0], MULL(c->band[0].scale_factor, qm4[ilow], 15));
+
+ // quantizer adaptation
+ c->band[0].log_factor = av_clip(MULL(c->band[0].log_factor, 127, 7) + qm3[ilow], 0, 18432);
+ c->band[0].scale_factor = scale(c->band[0].log_factor, 8);
+
+ if (avctx->sample_rate == 16000) {
+ int xout1 = 0, xout2 = 0, i;
+ const int dhigh = MULL(c->band[1].scale_factor, qm2[ihigh], 15);
+
+ // inverse adaptive quantizer
+ rhigh = av_clip(dhigh + c->band[1].s_predictor, -16384, 16383);
+
+ do_adaptive_prediction(&c->band[1], dhigh);
+
+ // quantizer adaptation
+ c->band[1].log_factor = av_clip(MULL(c->band[1].log_factor, 127, 7) + wh[ihigh&1],
+ 0, 22528);
+ c->band[1].scale_factor = scale(c->band[1].log_factor, 10);
+
+ // Apply the receive QMF
+ memmove(c->prev_samples, c->prev_samples + 2, 22*sizeof(c->prev_samples[0]));
+ c->prev_samples[22] = rlow + rhigh;
+ c->prev_samples[23] = rlow - rhigh;
+
+ for (i = 0; i < 12; i++) {
+ MAC16(xout2, c->prev_samples[2*i ], qmf_coeffs[i ]);
+ MAC16(xout1, c->prev_samples[2*i+1], qmf_coeffs[11-i]);
+ }
+ out_buf[out_len++] = av_clip_int16(xout1 >> 12);
+ out_buf[out_len++] = av_clip_int16(xout2 >> 12);
+ } else
+ out_buf[out_len++] = (int16_t) rlow;
+ }
+ *data_size = out_len << 1;
+ return buf_size;
+}
+
+
+static av_cold int g722_init(AVCodecContext * avctx)
+{
+ G722Context *c = (G722Context *) avctx->priv_data;
+
+ if (avctx->channels != 1) {
+ av_log(avctx, AV_LOG_ERROR, "Only mono tracks are allowed.\n");
+ return -1;
+ }
+
+ switch (avctx->bit_rate) {
+ case 64000:
+ case 56000:
+ case 48000:
+ c->bits_per_sample = avctx->bit_rate/8000;
+ break;
+ default:
+ av_log(avctx, AV_LOG_ERROR, "Unsupported bitrate [%d] only 48Kb, 56Kb and 64Kb are supported.\n", avctx->bit_rate);
+ return -1;
+ }
+
+ if (avctx->sample_rate != 8000 && avctx->sample_rate != 16000) {
+ av_log(avctx, AV_LOG_ERROR, "Unsupported sample rate [%d] only 8KHz and 16KHz are supported.\n", avctx->sample_rate);
+ return -1;
+ }
+
+ c->band[0].scale_factor = 32;
+ c->band[1].scale_factor = 8;
+
+ return 0;
+}
+
+AVCodec adpcm_g722_decoder = {
+ .name = "g722",
+ .type = CODEC_TYPE_AUDIO,
+ .id = CODEC_ID_ADPCM_G722,
+ .priv_data_size = sizeof(G722Context),
+ .init = g722_init,
+ .decode = g722_decode_frame,
+ .long_name = NULL_IF_CONFIG_SMALL("G.722 ADPCM"),
+};
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 3-add-g722_muxdemux.patch
Type: application/octet-stream
Size: 1317 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090321/bbffd22c/attachment.obj>
More information about the ffmpeg-devel
mailing list