#include "aac.h"
#include "aactab.h"
#include "aacenc.h"
+#include "aacenctab.h"
+#include "aacenc_utils.h"
#include "psymodel.h"
-#define AAC_MAX_CHANNELS 6
-
-#define ERROR_IF(cond, ...) \
- if (cond) { \
- av_log(avctx, AV_LOG_ERROR, __VA_ARGS__); \
- return AVERROR(EINVAL); \
- }
-
-#define WARN_IF(cond, ...) \
- if (cond) { \
- av_log(avctx, AV_LOG_WARNING, __VA_ARGS__); \
- }
-
-
-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,
- 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,
- swb_size_128_8
-};
-
-/** default channel configurations */
-static const uint8_t aac_chan_configs[6][5] = {
- {1, TYPE_SCE}, // 1 channel - single channel element
- {1, TYPE_CPE}, // 2 channels - channel pair
- {2, TYPE_SCE, TYPE_CPE}, // 3 channels - center + stereo
- {3, TYPE_SCE, TYPE_CPE, TYPE_SCE}, // 4 channels - front center + stereo + back center
- {3, TYPE_SCE, TYPE_CPE, TYPE_CPE}, // 5 channels - front center + stereo + back stereo
- {4, TYPE_SCE, TYPE_CPE, TYPE_CPE, TYPE_LFE}, // 6 channels - front center + stereo + back stereo + LFE
-};
-
-/**
- * Table to remap channels from libavcodec's default order to AAC order.
- */
-static const uint8_t aac_chan_maps[AAC_MAX_CHANNELS][AAC_MAX_CHANNELS] = {
- { 0 },
- { 0, 1 },
- { 2, 0, 1 },
- { 2, 0, 1, 3 },
- { 2, 0, 1, 3, 4 },
- { 2, 0, 1, 4, 5, 3 },
-};
-
/**
* Make AAC audio config object.
* @see 1.6.2.1 "Syntax - AudioSpecificConfig"
sce->ics.swb_sizes[i],
sce->sf_idx[w*16 + i],
sce->band_type[w*16 + i],
- s->lambda);
+ s->lambda, sce->ics.window_clipping[w]);
start += sce->ics.swb_sizes[i];
}
}
}
+/**
+ * Downscale spectral coefficients for near-clipping windows to avoid artifacts
+ */
+static void avoid_clipping(AACEncContext *s, SingleChannelElement *sce)
+{
+ int start, i, j, w;
+
+ if (sce->ics.clip_avoidance_factor < 1.0f) {
+ for (w = 0; w < sce->ics.num_windows; w++) {
+ start = 0;
+ for (i = 0; i < sce->ics.max_sfb; i++) {
+ float *swb_coeffs = sce->coeffs + start + w*128;
+ for (j = 0; j < sce->ics.swb_sizes[i]; j++)
+ swb_coeffs[j] *= sce->ics.clip_avoidance_factor;
+ start += sce->ics.swb_sizes[i];
+ }
+ }
+ }
+}
+
/**
* Encode one channel of audio data.
*/
for (ch = 0; ch < chans; ch++) {
IndividualChannelStream *ics = &cpe->ch[ch].ics;
int cur_channel = start_ch + ch;
+ float clip_avoidance_factor;
overlap = &samples[cur_channel][0];
samples2 = overlap + 1024;
la = samples2 + (448+64);
ics->num_windows = wi[ch].num_windows;
ics->swb_sizes = s->psy.bands [ics->num_windows == 8];
ics->num_swb = tag == TYPE_LFE ? ics->num_swb : s->psy.num_bands[ics->num_windows == 8];
+ clip_avoidance_factor = 0.0f;
for (w = 0; w < ics->num_windows; w++)
ics->group_len[w] = wi[ch].grouping[w];
+ for (w = 0; w < ics->num_windows; w++) {
+ if (wi[ch].clipping[w] > CLIP_AVOIDANCE_FACTOR) {
+ ics->window_clipping[w] = 1;
+ clip_avoidance_factor = FFMAX(clip_avoidance_factor, wi[ch].clipping[w]);
+ } else {
+ ics->window_clipping[w] = 0;
+ }
+ }
+ if (clip_avoidance_factor > CLIP_AVOIDANCE_FACTOR) {
+ ics->clip_avoidance_factor = CLIP_AVOIDANCE_FACTOR / clip_avoidance_factor;
+ } else {
+ ics->clip_avoidance_factor = 1.0f;
+ }
apply_window_and_mdct(s, &cpe->ch[ch], overlap);
if (isnan(cpe->ch->coeffs[0])) {
av_log(avctx, AV_LOG_ERROR, "Input contains NaN\n");
return AVERROR(EINVAL);
}
+ avoid_clipping(s, &cpe->ch[ch]);
}
start_ch += chans;
}
- if ((ret = ff_alloc_packet2(avctx, avpkt, 8192 * s->channels)) < 0)
+ if ((ret = ff_alloc_packet2(avctx, avpkt, 8192 * s->channels, 0)) < 0)
return ret;
do {
int frame_bits;
init_put_bits(&s->pb, avpkt->data, avpkt->size);
- if ((avctx->frame_number & 0xFF)==1 && !(avctx->flags & CODEC_FLAG_BITEXACT))
+ if ((avctx->frame_number & 0xFF)==1 && !(avctx->flags & AV_CODEC_FLAG_BITEXACT))
put_bitstream_info(s, LIBAVCODEC_IDENT);
start_ch = 0;
memset(chan_el_counter, 0, sizeof(chan_el_counter));
if (s->options.pns && s->coder->search_for_pns) {
for (ch = 0; ch < chans; ch++) {
s->cur_channel = start_ch + ch;
- s->coder->search_for_pns(s, avctx, &cpe->ch[ch], s->lambda);
+ s->coder->search_for_pns(s, avctx, &cpe->ch[ch]);
}
}
s->cur_channel = start_ch;
for (g = 0; g < ics->num_swb; g++)
cpe->ms_mask[w*16+g] = 1;
} else if (s->coder->search_for_ms) {
- s->coder->search_for_ms(s, cpe, s->lambda);
+ s->coder->search_for_ms(s, cpe);
}
}
if (chans > 1 && s->options.intensity_stereo && s->coder->search_for_is) {
- s->coder->search_for_is(s, avctx, cpe, s->lambda);
+ s->coder->search_for_is(s, avctx, cpe);
if (cpe->is_mode) is_mode = 1;
}
if (s->coder->set_special_band_scalefactors)
avctx->frame_bits = put_bits_count(&s->pb);
// rate control stuff
- if (!(avctx->flags & CODEC_FLAG_QSCALE)) {
+ if (!(avctx->flags & AV_CODEC_FLAG_QSCALE)) {
float ratio = avctx->bit_rate * 1024.0f / avctx->sample_rate / avctx->frame_bits;
s->lambda *= ratio;
s->lambda = FFMIN(s->lambda, 65536.f);
{
int ret = 0;
- s->fdsp = avpriv_float_dsp_alloc(avctx->flags & CODEC_FLAG_BITEXACT);
+ s->fdsp = avpriv_float_dsp_alloc(avctx->flags & AV_CODEC_FLAG_BITEXACT);
if (!s->fdsp)
return AVERROR(ENOMEM);
int ch;
FF_ALLOCZ_ARRAY_OR_GOTO(avctx, s->buffer.samples, s->channels, 3 * 1024 * sizeof(s->buffer.samples[0]), alloc_fail);
FF_ALLOCZ_ARRAY_OR_GOTO(avctx, s->cpe, s->chan_map[0], sizeof(ChannelElement), alloc_fail);
- FF_ALLOCZ_OR_GOTO(avctx, avctx->extradata, 5 + FF_INPUT_BUFFER_PADDING_SIZE, alloc_fail);
+ FF_ALLOCZ_OR_GOTO(avctx, avctx->extradata, 5 + AV_INPUT_BUFFER_PADDING_SIZE, alloc_fail);
for(ch = 0; ch < s->channels; ch++)
s->planar_samples[ch] = s->buffer.samples + 3 * 1024 * ch;
s->channels = avctx->channels;
- ERROR_IF(i == 16
- || i >= (sizeof(swb_size_1024) / sizeof(*swb_size_1024))
- || i >= (sizeof(swb_size_128) / sizeof(*swb_size_128)),
+ ERROR_IF(i == 16 || i >= swb_size_1024_len || i >= swb_size_128_len,
"Unsupported sample rate %d\n", avctx->sample_rate);
ERROR_IF(s->channels > AAC_MAX_CHANNELS,
"Unsupported number of channels: %d\n", s->channels);
{"auto", "Selected by the Encoder", 0, AV_OPT_TYPE_CONST, {.i64 = -1 }, INT_MIN, INT_MAX, AACENC_FLAGS, "stereo_mode"},
{"ms_off", "Disable Mid/Side coding", 0, AV_OPT_TYPE_CONST, {.i64 = 0 }, INT_MIN, INT_MAX, AACENC_FLAGS, "stereo_mode"},
{"ms_force", "Force Mid/Side for the whole frame if possible", 0, AV_OPT_TYPE_CONST, {.i64 = 1 }, INT_MIN, INT_MAX, AACENC_FLAGS, "stereo_mode"},
- {"aac_coder", "", offsetof(AACEncContext, options.aac_coder), AV_OPT_TYPE_INT, {.i64 = AAC_CODER_TWOLOOP}, 0, AAC_CODER_NB-1, AACENC_FLAGS, "aac_coder"},
+ {"aac_coder", "Coding algorithm", offsetof(AACEncContext, options.aac_coder), AV_OPT_TYPE_INT, {.i64 = AAC_CODER_TWOLOOP}, 0, AAC_CODER_NB-1, AACENC_FLAGS, "aac_coder"},
{"faac", "FAAC-inspired method", 0, AV_OPT_TYPE_CONST, {.i64 = AAC_CODER_FAAC}, INT_MIN, INT_MAX, AACENC_FLAGS, "aac_coder"},
{"anmr", "ANMR method", 0, AV_OPT_TYPE_CONST, {.i64 = AAC_CODER_ANMR}, INT_MIN, INT_MAX, AACENC_FLAGS, "aac_coder"},
{"twoloop", "Two loop searching method", 0, AV_OPT_TYPE_CONST, {.i64 = AAC_CODER_TWOLOOP}, INT_MIN, INT_MAX, AACENC_FLAGS, "aac_coder"},
LIBAVUTIL_VERSION_INT,
};
-/* duplicated from avpriv_mpeg4audio_sample_rates to avoid shared build
- * failures */
-static const int mpeg4audio_sample_rates[16] = {
- 96000, 88200, 64000, 48000, 44100, 32000,
- 24000, 22050, 16000, 12000, 11025, 8000, 7350
-};
-
AVCodec ff_aac_encoder = {
.name = "aac",
.long_name = NULL_IF_CONFIG_SMALL("AAC (Advanced Audio Coding)"),
.encode2 = aac_encode_frame,
.close = aac_encode_end,
.supported_samplerates = mpeg4audio_sample_rates,
- .capabilities = CODEC_CAP_SMALL_LAST_FRAME | CODEC_CAP_DELAY |
- CODEC_CAP_EXPERIMENTAL,
+ .capabilities = AV_CODEC_CAP_SMALL_LAST_FRAME | AV_CODEC_CAP_DELAY |
+ AV_CODEC_CAP_EXPERIMENTAL,
.sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_FLTP,
AV_SAMPLE_FMT_NONE },
.priv_class = &aacenc_class,