diff options
author | Martin Storsjö <martin@martin.st> | 2010-09-09 19:21:16 +0000 |
---|---|---|
committer | Martin Storsjö <martin@martin.st> | 2010-09-09 19:21:16 +0000 |
commit | 58e37eafce10bbcf10b111cafb7ba8586c76ef4b (patch) | |
tree | 54bd919cd566792fc1acb80c4cc06425efec892d | |
parent | a3f0d2b9674b4bbbdb6eb28a1e600ba324a5853b (diff) | |
download | ffmpeg-streaming-58e37eafce10bbcf10b111cafb7ba8586c76ef4b.zip ffmpeg-streaming-58e37eafce10bbcf10b111cafb7ba8586c76ef4b.tar.gz |
Add G.722 ADPCM audio decoder
Originally committed as revision 25086 to svn://svn.ffmpeg.org/ffmpeg/trunk
-rw-r--r-- | Changelog | 1 | ||||
-rw-r--r-- | doc/general.texi | 1 | ||||
-rw-r--r-- | libavcodec/Makefile | 1 | ||||
-rw-r--r-- | libavcodec/allcodecs.c | 1 | ||||
-rw-r--r-- | libavcodec/avcodec.h | 5 | ||||
-rw-r--r-- | libavcodec/g722.c | 304 |
6 files changed, 311 insertions, 2 deletions
@@ -33,6 +33,7 @@ version <next>: - Apple HTTP Live Streaming demuxer - a64 codec - MMS-HTTP support +- G.722 ADPCM audio decoder version 0.6: diff --git a/doc/general.texi b/doc/general.texi index b775e68..a692e04 100644 --- a/doc/general.texi +++ b/doc/general.texi @@ -535,6 +535,7 @@ following image formats are supported: @item ADPCM Electronic Arts R2 @tab @tab X @item ADPCM Electronic Arts R3 @tab @tab X @item ADPCM Electronic Arts XAS @tab @tab X +@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 diff --git a/libavcodec/Makefile b/libavcodec/Makefile index a59208f..e0d7028 100644 --- a/libavcodec/Makefile +++ b/libavcodec/Makefile @@ -475,6 +475,7 @@ OBJS-$(CONFIG_ADPCM_EA_R1_DECODER) += adpcm.o OBJS-$(CONFIG_ADPCM_EA_R2_DECODER) += adpcm.o OBJS-$(CONFIG_ADPCM_EA_R3_DECODER) += adpcm.o OBJS-$(CONFIG_ADPCM_EA_XAS_DECODER) += adpcm.o +OBJS-$(CONFIG_ADPCM_G722_DECODER) += g722.o OBJS-$(CONFIG_ADPCM_G726_DECODER) += g726.o OBJS-$(CONFIG_ADPCM_G726_ENCODER) += g726.o OBJS-$(CONFIG_ADPCM_IMA_AMV_DECODER) += adpcm.o diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c index bf28028..31cfadd 100644 --- a/libavcodec/allcodecs.c +++ b/libavcodec/allcodecs.c @@ -317,6 +317,7 @@ void avcodec_register_all(void) REGISTER_DECODER (ADPCM_EA_R2, adpcm_ea_r2); REGISTER_DECODER (ADPCM_EA_R3, adpcm_ea_r3); REGISTER_DECODER (ADPCM_EA_XAS, adpcm_ea_xas); + REGISTER_DECODER (ADPCM_G722, adpcm_g722); REGISTER_ENCDEC (ADPCM_G726, adpcm_g726); REGISTER_DECODER (ADPCM_IMA_AMV, adpcm_ima_amv); REGISTER_DECODER (ADPCM_IMA_DK3, adpcm_ima_dk3); diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h index 84ab1b4..484880e 100644 --- a/libavcodec/avcodec.h +++ b/libavcodec/avcodec.h @@ -31,8 +31,8 @@ #include "libavutil/cpu.h" #define LIBAVCODEC_VERSION_MAJOR 52 -#define LIBAVCODEC_VERSION_MINOR 87 -#define LIBAVCODEC_VERSION_MICRO 5 +#define LIBAVCODEC_VERSION_MINOR 88 +#define LIBAVCODEC_VERSION_MICRO 0 #define LIBAVCODEC_VERSION_INT AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \ LIBAVCODEC_VERSION_MINOR, \ @@ -284,6 +284,7 @@ enum CodecID { 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, diff --git a/libavcodec/g722.c b/libavcodec/g722.c new file mode 100644 index 0000000..8707d16 --- /dev/null +++ b/libavcodec/g722.c @@ -0,0 +1,304 @@ +/* + * G.722 ADPCM audio decoder + * + * Copyright (c) CMU 1993 Computer Science, Speech Group + * Chengxiang Lu and Alex Hauptmann + * Copyright (c) 2005 Steve Underwood <steveu at coppice.org> + * Copyright (c) 2009 Kenan Gillet + * Copyright (c) 2010 Martin Storsjo + * + * 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 + * + * G.722 ADPCM audio codec + * + * This G.722 decoder is a bit-exact implementation of the ITU G.722 + * specification for all three specified bitrates - 64000bps, 56000bps + * and 48000bps. It passes the ITU tests. + * + * @note For the 56000bps and 48000bps bitrates, the lowest 1 or 2 bits + * respectively of each byte are ignored. + */ + +#include "avcodec.h" +#include "mathops.h" +#include "get_bits.h" + +#define PREV_SAMPLES_BUF_SIZE 1024 + +typedef struct { + int16_t prev_samples[PREV_SAMPLES_BUF_SIZE]; ///< memory of past decoded samples + int prev_samples_pos; ///< the number of values in prev_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; ///< previous output signal from zero predictor + int8_t part_reconst_mem[2]; ///< signs of previous partially reconstructed signals + int16_t prev_qtzd_reconst; ///< previous quantized reconstructed signal (internal value, using low_inv_quant4) + int16_t pole_mem[2]; ///< second-order pole section coefficient buffer + int32_t diff_mem[6]; ///< quantizer difference signal memory + int16_t zero_mem[6]; ///< Seventh-order zero section coefficient buffer + int16_t log_factor; ///< delayed 2-logarithmic quantizer factor + int16_t scale_factor; ///< delayed quantizer scale factor + } band[2]; +} G722Context; + + +static const int8_t sign_lookup[2] = { -1, 1 }; + +static const int16_t inv_log2_table[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 high_log_factor_step[2] = { 798, -214 }; +static const int16_t high_inv_quant[4] = { -926, -202, 926, 202 }; +/** + * low_log_factor_step[index] == wl[rl42[index]] + */ +static const int16_t low_log_factor_step[16] = { + -60, 3042, 1198, 538, 334, 172, 58, -30, + 3042, 1198, 538, 334, 172, 58, -30, -60 +}; +static const int16_t low_inv_quant4[16] = { + 0, -2557, -1612, -1121, -786, -530, -323, -150, + 2557, 1612, 1121, 786, 530, 323, 150, 0 +}; + +/** + * quadrature mirror filter (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, +}; + + +/** + * adaptive predictor + * + * @param cur_diff the dequantized and scaled delta calculated from the + * current codeword + */ +static void do_adaptive_prediction(struct G722Band *band, const int cur_diff) +{ + int sg[2], limit, i, cur_qtzd_reconst; + + const int 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) + (band->pole_mem[1] * 127 >> 7), -12288, 12288); + + limit = 15360 - band->pole_mem[1]; + band->pole_mem[0] = av_clip(-192 * sg[0] + (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]^cur_diff) < 0 ? -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] = av_clip_int16(cur_diff << 1); + + band->s_zero = 0; + for (i = 5; i >= 0; i--) + band->s_zero += (band->zero_mem[i]*band->diff_mem[i]) >> 15; + + + cur_qtzd_reconst = av_clip_int16((band->s_predictor + cur_diff) << 1); + band->s_predictor = av_clip_int16(band->s_zero + + (band->pole_mem[0] * cur_qtzd_reconst >> 15) + + (band->pole_mem[1] * band->prev_qtzd_reconst >> 15)); + band->prev_qtzd_reconst = cur_qtzd_reconst; +} + +static int inline linear_scale_factor(const int log_factor) +{ + const int wd1 = inv_log2_table[(log_factor >> 6) & 31]; + const int shift = log_factor >> 11; + return shift < 0 ? wd1 >> -shift : wd1 << shift; +} + +static void update_low_predictor(struct G722Band *band, const int ilow) +{ + do_adaptive_prediction(band, + band->scale_factor * low_inv_quant4[ilow] >> 10); + + // quantizer adaptation + band->log_factor = av_clip((band->log_factor * 127 >> 7) + + low_log_factor_step[ilow], 0, 18432); + band->scale_factor = linear_scale_factor(band->log_factor - (8 << 11)); +} + +static void update_high_predictor(struct G722Band *band, const int dhigh, + const int ihigh) +{ + do_adaptive_prediction(band, dhigh); + + // quantizer adaptation + band->log_factor = av_clip((band->log_factor * 127 >> 7) + + high_log_factor_step[ihigh&1], 0, 22528); + band->scale_factor = linear_scale_factor(band->log_factor - (10 << 11)); +} + +static void apply_qmf(const int16_t *prev_samples, int *xout1, int *xout2) +{ + int i; + + *xout1 = 0; + *xout2 = 0; + for (i = 0; i < 12; i++) { + MAC16(*xout2, prev_samples[2*i ], qmf_coeffs[i ]); + MAC16(*xout1, prev_samples[2*i+1], qmf_coeffs[11-i]); + } +} + +static av_cold int g722_init(AVCodecContext * avctx) +{ + G722Context *c = avctx->priv_data; + + if (avctx->channels != 1) { + av_log(avctx, AV_LOG_ERROR, "Only mono tracks are allowed.\n"); + return AVERROR_INVALIDDATA; + } + avctx->sample_fmt = SAMPLE_FMT_S16; + + switch (avctx->bits_per_coded_sample) { + case 8: + case 7: + case 6: + break; + default: + av_log(avctx, AV_LOG_WARNING, "Unsupported bits_per_coded_sample [%d], " + "assuming 8\n", + avctx->bits_per_coded_sample); + case 0: + avctx->bits_per_coded_sample = 8; + break; + } + + c->band[0].scale_factor = 8; + c->band[1].scale_factor = 2; + c->prev_samples_pos = 22; + + if (avctx->lowres) + avctx->sample_rate /= 2; + + return 0; +} + +static const int16_t low_inv_quant5[32] = { + -35, -35, -2919, -2195, -1765, -1458, -1219, -1023, + -858, -714, -587, -473, -370, -276, -190, -110, + 2919, 2195, 1765, 1458, 1219, 1023, 858, 714, + 587, 473, 370, 276, 190, 110, 35, -35 +}; +static const int16_t low_inv_quant6[64] = { + -17, -17, -17, -17, -3101, -2738, -2376, -2088, + -1873, -1689, -1535, -1399, -1279, -1170, -1072, -982, + -899, -822, -750, -682, -618, -558, -501, -447, + -396, -347, -300, -254, -211, -170, -130, -91, + 3101, 2738, 2376, 2088, 1873, 1689, 1535, 1399, + 1279, 1170, 1072, 982, 899, 822, 750, 682, + 618, 558, 501, 447, 396, 347, 300, 254, + 211, 170, 130, 91, 54, 17, -54, -17 +}; + +static const int16_t *low_inv_quants[3] = { low_inv_quant6, low_inv_quant5, + low_inv_quant4 }; + +static int g722_decode_frame(AVCodecContext *avctx, void *data, + int *data_size, AVPacket *avpkt) +{ + G722Context *c = avctx->priv_data; + int16_t *out_buf = data; + int j, out_len = 0; + const int skip = 8 - avctx->bits_per_coded_sample; + const int16_t *quantizer_table = low_inv_quants[skip]; + GetBitContext gb; + + init_get_bits(&gb, avpkt->data, avpkt->size * 8); + + for (j = 0; j < avpkt->size; j++) { + int ilow, ihigh, rlow; + + ihigh = get_bits(&gb, 2); + ilow = get_bits(&gb, 6 - skip); + skip_bits(&gb, skip); + + rlow = av_clip((c->band[0].scale_factor * quantizer_table[ilow] >> 10) + + c->band[0].s_predictor, -16384, 16383); + + update_low_predictor(&c->band[0], ilow >> (2 - skip)); + + if (!avctx->lowres) { + const int dhigh = c->band[1].scale_factor * + high_inv_quant[ihigh] >> 10; + const int rhigh = av_clip(dhigh + c->band[1].s_predictor, + -16384, 16383); + int xout1, xout2; + + update_high_predictor(&c->band[1], dhigh, ihigh); + + c->prev_samples[c->prev_samples_pos++] = rlow + rhigh; + c->prev_samples[c->prev_samples_pos++] = rlow - rhigh; + apply_qmf(c->prev_samples + c->prev_samples_pos - 24, + &xout1, &xout2); + out_buf[out_len++] = av_clip_int16(xout1 >> 12); + out_buf[out_len++] = av_clip_int16(xout2 >> 12); + if (c->prev_samples_pos >= PREV_SAMPLES_BUF_SIZE) { + memmove(c->prev_samples, + c->prev_samples + c->prev_samples_pos - 22, + 22 * sizeof(c->prev_samples[0])); + c->prev_samples_pos = 22; + } + } else + out_buf[out_len++] = rlow; + } + *data_size = out_len << 1; + return avpkt->size; +} + +AVCodec adpcm_g722_decoder = { + .name = "g722", + .type = AVMEDIA_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"), + .max_lowres = 1, +}; + |