Mercurial > libavcodec.hg
changeset 7571:58f6bb760994 libavcodec
Okayed parts of AAC encoder
author | kostya |
---|---|
date | Thu, 14 Aug 2008 05:52:29 +0000 |
parents | bb32f8a4c0f9 |
children | d72f9295fe6e |
files | aacenc.c |
diffstat | 1 files changed, 313 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/aacenc.c Thu Aug 14 05:52:29 2008 +0000 @@ -0,0 +1,313 @@ +/* + * AAC encoder + * Copyright (C) 2008 Konstantin Shishkov + * + * 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 aacenc.c + * AAC encoder + */ + +/*********************************** + * TODOs: + * psy model selection with some option + * change greedy codebook search into something more optimal, like Viterbi algorithm + * determine run lengths along with codebook + ***********************************/ + +#include "avcodec.h" +#include "bitstream.h" +#include "dsputil.h" +#include "mpeg4audio.h" + +#include "aacpsy.h" +#include "aac.h" +#include "aactab.h" + +static const uint8_t swb_size_1024_96[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, 8, + 12, 12, 12, 12, 12, 16, 16, 24, 28, 36, 44, + 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64 +}; + +static const uint8_t swb_size_1024_64[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, + 12, 12, 12, 16, 16, 16, 20, 24, 24, 28, 36, + 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40 +}; + +static const uint8_t swb_size_1024_48[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, 8, 8, 8, + 12, 12, 12, 12, 16, 16, 20, 20, 24, 24, 28, 28, + 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, + 96 +}; + +static const uint8_t swb_size_1024_32[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, 8, 8, 8, + 12, 12, 12, 12, 16, 16, 20, 20, 24, 24, 28, 28, + 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32, 32 +}; + +static const uint8_t swb_size_1024_24[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 12, 12, 12, 12, 16, 16, 16, 20, 20, 24, 24, 28, 28, + 32, 36, 36, 40, 44, 48, 52, 52, 64, 64, 64, 64, 64 +}; + +static const uint8_t swb_size_1024_16[] = { + 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, + 12, 12, 12, 12, 12, 12, 12, 12, 12, 16, 16, 16, 16, 20, 20, 20, 24, 24, 28, 28, + 32, 36, 40, 40, 44, 48, 52, 56, 60, 64, 64, 64 +}; + +static const uint8_t swb_size_1024_8[] = { + 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, + 16, 16, 16, 16, 16, 16, 16, 20, 20, 20, 20, 24, 24, 24, 28, 28, + 32, 36, 36, 40, 44, 48, 52, 56, 60, 64, 80 +}; + +static const uint8_t *swb_size_1024[] = { + swb_size_1024_96, swb_size_1024_96, swb_size_1024_64, + swb_size_1024_48, swb_size_1024_48, swb_size_1024_32, + swb_size_1024_24, swb_size_1024_24, swb_size_1024_16, + swb_size_1024_16, swb_size_1024_16, swb_size_1024_8 +}; + +static const uint8_t swb_size_128_96[] = { + 4, 4, 4, 4, 4, 4, 8, 8, 8, 16, 28, 36 +}; + +static const uint8_t swb_size_128_48[] = { + 4, 4, 4, 4, 4, 8, 8, 8, 12, 12, 12, 16, 16, 16 +}; + +static const uint8_t swb_size_128_24[] = { + 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 12, 12, 16, 16, 20 +}; + +static const uint8_t swb_size_128_16[] = { + 4, 4, 4, 4, 4, 4, 4, 4, 8, 8, 12, 12, 16, 20, 20 +}; + +static const uint8_t swb_size_128_8[] = { + 4, 4, 4, 4, 4, 4, 4, 8, 8, 8, 8, 12, 16, 20, 20 +}; + +static const uint8_t *swb_size_128[] = { + /* the last entry on the following row is swb_size_128_64 but is a + duplicate of swb_size_128_96 */ + swb_size_128_96, swb_size_128_96, swb_size_128_96, + swb_size_128_48, swb_size_128_48, swb_size_128_48, + swb_size_128_24, swb_size_128_24, swb_size_128_16, + swb_size_128_16, swb_size_128_16, swb_size_128_8 +}; + +#define CB_UNSIGNED 0x01 ///< coefficients are coded as absolute values +#define CB_PAIRS 0x02 ///< coefficients are grouped into pairs before coding (quads by default) +#define CB_ESCAPE 0x04 ///< codebook allows escapes + +/** spectral coefficients codebook information */ +static const struct { + int16_t maxval; ///< maximum possible value + int8_t cb_num; ///< codebook number + uint8_t flags; ///< codebook features +} aac_cb_info[] = { + { 0, -1, CB_UNSIGNED }, // zero codebook + { 1, 0, 0 }, + { 1, 1, 0 }, + { 2, 2, CB_UNSIGNED }, + { 2, 3, CB_UNSIGNED }, + { 4, 4, CB_PAIRS }, + { 4, 5, CB_PAIRS }, + { 7, 6, CB_PAIRS | CB_UNSIGNED }, + { 7, 7, CB_PAIRS | CB_UNSIGNED }, + { 12, 8, CB_PAIRS | CB_UNSIGNED }, + { 12, 9, CB_PAIRS | CB_UNSIGNED }, + { 8191, 10, CB_PAIRS | CB_UNSIGNED | CB_ESCAPE }, + { -1, -1, 0 }, // reserved + { -1, -1, 0 }, // perceptual noise substitution + { -1, -1, 0 }, // intensity out-of-phase + { -1, -1, 0 }, // intensity in-phase +}; + +/** default channel configurations */ +static const uint8_t aac_chan_configs[6][5] = { + {1, ID_SCE}, // 1 channel - single channel element + {1, ID_CPE}, // 2 channels - channel pair + {2, ID_SCE, ID_CPE}, // 3 channels - center + stereo + {3, ID_SCE, ID_CPE, ID_SCE}, // 4 channels - front center + stereo + back center + {3, ID_SCE, ID_CPE, ID_CPE}, // 5 channels - front center + stereo + back stereo + {4, ID_SCE, ID_CPE, ID_CPE, ID_LFE}, // 6 channels - front center + stereo + back stereo + LFE +}; + +/** + * AAC encoder context + */ +typedef struct { + PutBitContext pb; + MDCTContext mdct1024; ///< long (1024 samples) frame transform context + MDCTContext mdct128; ///< short (128 samples) frame transform context + DSPContext dsp; + DECLARE_ALIGNED_16(FFTSample, output[2048]); ///< temporary buffer for MDCT input coefficients + DECLARE_ALIGNED_16(FFTSample, tmp[1024]); ///< temporary buffer used by MDCT + int16_t* samples; ///< saved preprocessed input + + int samplerate_index; ///< MPEG-4 samplerate index + const uint8_t *swb_sizes1024; ///< scalefactor band sizes for long frame + int swb_num1024; ///< number of scalefactor bands for long frame + const uint8_t *swb_sizes128; ///< scalefactor band sizes for short frame + int swb_num128; ///< number of scalefactor bands for short frame + + ChannelElement *cpe; ///< channel elements + AACPsyContext psy; ///< psychoacoustic model context + int last_frame; +} AACEncContext; + +/** + * Make AAC audio config object. + * @see 1.6.2.1 "Syntax - AudioSpecificConfig" + */ +static void put_audio_specific_config(AVCodecContext *avctx) +{ + PutBitContext pb; + AACEncContext *s = avctx->priv_data; + + init_put_bits(&pb, avctx->extradata, avctx->extradata_size*8); + put_bits(&pb, 5, 2); //object type - AAC-LC + put_bits(&pb, 4, s->samplerate_index); //sample rate index + put_bits(&pb, 4, avctx->channels); + //GASpecificConfig + put_bits(&pb, 1, 0); //frame length - 1024 samples + put_bits(&pb, 1, 0); //does not depend on core coder + put_bits(&pb, 1, 0); //is not extension + flush_put_bits(&pb); +} + +static av_cold int aac_encode_init(AVCodecContext *avctx) +{ + AACEncContext *s = avctx->priv_data; + int i; + + avctx->frame_size = 1024; + + for(i = 0; i < 16; i++) + if(avctx->sample_rate == ff_mpeg4audio_sample_rates[i]) + break; + if(i == 16){ + av_log(avctx, AV_LOG_ERROR, "Unsupported sample rate %d\n", avctx->sample_rate); + return -1; + } + if(avctx->channels > 6){ + av_log(avctx, AV_LOG_ERROR, "Unsupported number of channels: %d\n", avctx->channels); + return -1; + } + s->samplerate_index = i; + s->swb_sizes1024 = swb_size_1024[i]; + s->swb_num1024 = ff_aac_num_swb_1024[i]; + s->swb_sizes128 = swb_size_128[i]; + s->swb_num128 = ff_aac_num_swb_128[i]; + + dsputil_init(&s->dsp, avctx); + ff_mdct_init(&s->mdct1024, 11, 0); + ff_mdct_init(&s->mdct128, 8, 0); + // window init + ff_kbd_window_init(ff_aac_kbd_long_1024, 4.0, 1024); + ff_kbd_window_init(ff_aac_kbd_short_128, 6.0, 128); + ff_sine_window_init(ff_aac_sine_long_1024, 1024); + ff_sine_window_init(ff_aac_sine_short_128, 128); + + s->samples = av_malloc(2 * 1024 * avctx->channels * sizeof(s->samples[0])); + s->cpe = av_mallocz(sizeof(ChannelElement) * aac_chan_configs[avctx->channels-1][0]); + if(ff_aac_psy_init(&s->psy, avctx, AAC_PSY_3GPP, aac_chan_configs[avctx->channels-1][0], 0, s->swb_sizes1024, s->swb_num1024, s->swb_sizes128, s->swb_num128) < 0){ + av_log(avctx, AV_LOG_ERROR, "Cannot initialize selected model.\n"); + return -1; + } + avctx->extradata = av_malloc(2); + avctx->extradata_size = 2; + put_audio_specific_config(avctx); + return 0; +} + +/** + * Encode ics_info element. + * @see Table 4.6 (syntax of ics_info) + */ +static void put_ics_info(AVCodecContext *avctx, IndividualChannelStream *info) +{ + AACEncContext *s = avctx->priv_data; + int i; + + put_bits(&s->pb, 1, 0); // ics_reserved bit + put_bits(&s->pb, 2, info->window_sequence[0]); + put_bits(&s->pb, 1, info->use_kb_window[0]); + if(info->window_sequence[0] != EIGHT_SHORT_SEQUENCE){ + put_bits(&s->pb, 6, info->max_sfb); + put_bits(&s->pb, 1, 0); // no prediction + }else{ + put_bits(&s->pb, 4, info->max_sfb); + for(i = 1; i < info->num_windows; i++) + put_bits(&s->pb, 1, info->group_len[i]); + } +} + +/** + * Write some auxiliary information about the created AAC file. + */ +static void put_bitstream_info(AVCodecContext *avctx, AACEncContext *s, const char *name) +{ + int i, namelen, padbits; + + namelen = strlen(name) + 2; + put_bits(&s->pb, 3, ID_FIL); + put_bits(&s->pb, 4, FFMIN(namelen, 15)); + if(namelen >= 15) + put_bits(&s->pb, 8, namelen - 16); + put_bits(&s->pb, 4, 0); //extension type - filler + padbits = 8 - (put_bits_count(&s->pb) & 7); + align_put_bits(&s->pb); + for(i = 0; i < namelen - 2; i++) + put_bits(&s->pb, 8, name[i]); + put_bits(&s->pb, 12 - padbits, 0); +} + +static av_cold int aac_encode_end(AVCodecContext *avctx) +{ + AACEncContext *s = avctx->priv_data; + + ff_mdct_end(&s->mdct1024); + ff_mdct_end(&s->mdct128); + ff_aac_psy_end(&s->psy); + av_freep(&s->samples); + av_freep(&s->cpe); + return 0; +} + +AVCodec aac_encoder = { + "aac", + CODEC_TYPE_AUDIO, + CODEC_ID_AAC, + sizeof(AACEncContext), + aac_encode_init, + aac_encode_frame, + aac_encode_end, + .capabilities = CODEC_CAP_SMALL_LAST_FRAME | CODEC_CAP_DELAY, + .sample_fmts = (enum SampleFormat[]){SAMPLE_FMT_S16,SAMPLE_FMT_NONE}, + .long_name = NULL_IF_CONFIG_SMALL("Advanced Audio Coding"), +};