Go to the documentation of this file.
102 enum OCStatus oc_type,
int get_new_frame);
104 #define overread_err "Input buffer exhausted before END element found\n"
109 for (
i = 0;
i < tags;
i++) {
112 sum += (1 + (syn_ele ==
TYPE_CPE)) *
206 uint8_t (*layout_map)[3],
int offset, uint64_t
left,
207 uint64_t right,
int pos, uint64_t *
layout)
211 .av_position =
left | right,
213 .elem_id = layout_map[
offset][1],
216 if (e2c_vec[
offset].av_position != UINT64_MAX)
224 .elem_id = layout_map[
offset][1],
228 .av_position = right,
230 .elem_id = layout_map[
offset + 1][1],
233 if (
left != UINT64_MAX)
236 if (right != UINT64_MAX)
246 int num_pos_channels = 0;
250 for (
i = current;
i < tags;
i++) {
251 if (layout_map[
i][2] !=
pos)
261 num_pos_channels += 2;
272 return num_pos_channels;
276 uint64_t *
layout,
int tags,
int layer,
int pos,
int *current)
278 int i = *current, j = 0;
281 if (nb_channels < 0 || nb_channels > 5)
285 while (nb_channels) {
290 .syn_ele = layout_map[
i][0],
291 .elem_id = layout_map[
i][1],
294 *
layout |= e2c_vec[
i].av_position;
304 while (nb_channels & 1) {
311 .syn_ele = layout_map[
i][0],
312 .elem_id = layout_map[
i][1],
315 *
layout |= e2c_vec[
i].av_position;
321 while (nb_channels >= 2) {
332 while (nb_channels & 1) {
337 .syn_ele = layout_map[
i][0],
338 .elem_id = layout_map[
i][1],
341 *
layout |= e2c_vec[
i].av_position;
355 int i, n, total_non_cc_elements;
362 for (n = 0,
i = 0; n < 3 &&
i < tags; n++) {
377 total_non_cc_elements = n =
i;
395 for (
i = 1;
i < n;
i++)
405 for (
i = 0;
i < total_non_cc_elements;
i++) {
421 ac->
oc[0] = ac->
oc[1];
434 ac->
oc[1] = ac->
oc[0];
449 enum OCStatus oc_type,
int get_new_frame)
455 uint8_t type_counts[
TYPE_END] = { 0 };
458 memcpy(ac->
oc[1].
layout_map, layout_map, tags *
sizeof(layout_map[0]));
461 for (
i = 0;
i < tags;
i++) {
462 int type = layout_map[
i][0];
463 int id = layout_map[
i][1];
472 #if FF_API_OLD_CHANNEL_LAYOUT
481 for (
i = 0;
i < tags;
i++) {
482 int type = layout_map[
i][0];
483 int id = layout_map[
i][1];
484 int iid = id_map[
type][
id];
485 int position = layout_map[
i][2];
529 for (j = 0; j <= 1; j++) {
544 uint8_t (*layout_map)[3],
548 if (channel_config < 1 || (channel_config > 7 && channel_config < 11) ||
549 channel_config > 14) {
551 "invalid default channel configuration (%d)\n",
557 *tags *
sizeof(*layout_map));
575 " instead of a spec-compliant 7.1(wide) layout, use -strict %d to decode"
600 &layout_map_tags, 2) < 0)
619 layout_map[0][0] = layout_map[1][0] =
TYPE_SCE;
621 layout_map[0][1] = 0;
622 layout_map[1][1] = 1;
667 "This stream seems to incorrectly report its last channel as %s[%d], mapping to LFE[0]\n",
689 "This stream seems to incorrectly report its last channel as %s[%d], mapping to SCE[1]\n",
754 layout_map[0][2] =
type;
760 int reference_position) {
772 uint8_t (*layout_map)[3],
775 int num_front, num_side, num_back, num_lfe, num_assoc_data, num_cc;
785 "Sample rate index in program config element does not "
786 "match the sample rate index configured by the container.\n");
803 if (
get_bits_left(gb) < 5 * (num_front + num_side + num_back + num_cc) + 4 *(num_lfe + num_assoc_data + num_cc)) {
843 int get_bit_alignment,
847 int extension_flag,
ret, ep_config, res_flags;
867 if (channel_config == 0) {
869 tags =
decode_pce(avctx, m4ac, layout_map, gb, get_bit_alignment);
874 &tags, channel_config)))
880 }
else if (m4ac->
sbr == 1 && m4ac->
ps == -1)
886 if (extension_flag) {
899 "AAC data resilience (flags %x)",
915 "epConfig %d", ep_config);
927 int ret, ep_config, res_flags;
930 const int ELDEXT_TERM = 0;
939 "AAC data resilience (flags %x)",
950 while (
get_bits(gb, 4) != ELDEXT_TERM) {
964 &tags, channel_config)))
973 "epConfig %d", ep_config);
995 int get_bit_alignment,
1009 "invalid sampling rate index %d\n",
1017 "invalid low delay sampling rate index %d\n",
1043 "Audio object type %s%d",
1044 m4ac->
sbr == 1 ?
"SBR+" :
"",
1050 "AOT %d chan config %d sampling index %d (%d) SBR %d PS %d\n",
1061 const uint8_t *
data, int64_t bit_size,
1067 if (bit_size < 0 || bit_size > INT_MAX) {
1072 ff_dlog(avctx,
"audio specific config size %d\n", (
int)bit_size >> 3);
1073 for (
i = 0; i < bit_size >> 3;
i++)
1093 union {
unsigned u;
int s; } v = { previous_val * 1664525
u + 1013904223 };
1106 if (92017 <= rate)
return 0;
1107 else if (75132 <= rate)
return 1;
1108 else if (55426 <= rate)
return 2;
1109 else if (46009 <= rate)
return 3;
1110 else if (37566 <= rate)
return 4;
1111 else if (27713 <= rate)
return 5;
1112 else if (23004 <= rate)
return 6;
1113 else if (18783 <= rate)
return 7;
1114 else if (13856 <= rate)
return 8;
1115 else if (11502 <= rate)
return 9;
1116 else if (9391 <= rate)
return 10;
1132 294 + 306 + 268 + 510 + 366 + 462];
1133 for (
unsigned i = 0,
offset = 0;
i < 11;
i++) {
1213 int layout_map_tags;
1256 #define MDCT_INIT(s, fn, len, sval) \
1258 ret = av_tx_init(&s, &fn, TX_TYPE, 1, len, &scale, 0); \
1308 "Invalid Predictor Reset Group.\n");
1354 "AAC LD is only defined for ONLY_LONG_SEQUENCE but "
1367 for (
i = 0;
i < 7;
i++) {
1424 "Prediction is not allowed in AAC-LC.\n");
1429 "LTP in ER AAC LD not yet implemented.\n");
1441 "Number of scalefactor bands in group (%d) "
1442 "exceeds limit (%d).\n",
1469 while (k < ics->max_sfb) {
1470 uint8_t sect_end = k;
1472 int sect_band_type =
get_bits(gb, 4);
1473 if (sect_band_type == 12) {
1479 sect_end += sect_len_incr;
1484 if (sect_end > ics->
max_sfb) {
1486 "Number of bands (%d) exceeds limit (%d).\n",
1490 }
while (sect_len_incr == (1 <<
bits) - 1);
1491 for (; k < sect_end; k++) {
1492 band_type [idx] = sect_band_type;
1493 band_type_run_end[idx++] = sect_end;
1511 unsigned int global_gain,
1514 int band_type_run_end[120])
1522 int run_end = band_type_run_end[idx];
1523 if (band_type[idx] ==
ZERO_BT) {
1524 for (;
i < run_end;
i++, idx++)
1528 for (;
i < run_end;
i++, idx++) {
1531 if (
offset[2] != clipped_offset) {
1533 "If you heard an audible artifact, there may be a bug in the decoder. "
1534 "Clipped intensity stereo position (%d -> %d)",
1535 offset[2], clipped_offset);
1538 sf[idx] = 100 - clipped_offset;
1543 }
else if (band_type[idx] ==
NOISE_BT) {
1544 for (;
i < run_end;
i++, idx++) {
1545 if (noise_flag-- > 0)
1550 if (
offset[1] != clipped_offset) {
1552 "If you heard an audible artifact, there may be a bug in the decoder. "
1553 "Clipped noise gain (%d -> %d)",
1554 offset[1], clipped_offset);
1557 sf[idx] = -(100 + clipped_offset);
1563 for (;
i < run_end;
i++, idx++) {
1567 "Scalefactor (%d) out of range.\n",
offset[0]);
1586 const uint16_t *swb_offset,
int num_swb)
1591 if (pulse_swb >= num_swb)
1593 pulse->
pos[0] = swb_offset[pulse_swb];
1595 if (pulse->
pos[0] >= swb_offset[num_swb])
1600 if (pulse->
pos[
i] >= swb_offset[num_swb])
1615 int w,
filt,
i, coef_len, coef_res, coef_compress;
1628 "TNS filter order %d is greater than maximum %d.\n",
1636 coef_len = coef_res + 3 - coef_compress;
1637 tmp2_idx = 2 * coef_compress + coef_res;
1660 if (ms_present == 1) {
1661 for (idx = 0; idx < max_idx; idx++)
1663 }
else if (ms_present == 2) {
1682 int pulse_present,
const Pulse *pulse,
1686 int i, k,
g, idx = 0;
1699 const unsigned cbt_m1 = band_type[idx] - 1;
1705 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1706 memset(cfo, 0, off_len *
sizeof(*cfo));
1708 }
else if (cbt_m1 ==
NOISE_BT - 1) {
1709 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1712 for (k = 0; k < off_len; k++) {
1717 band_energy = ac->
fdsp->scalarproduct_fixed(cfo, cfo, off_len);
1723 for (k = 0; k < off_len; k++) {
1740 switch (cbt_m1 >> 1) {
1742 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1756 cf =
VMUL4(cf, vq, cb_idx, sf + idx);
1763 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1776 nnz = cb_idx >> 8 & 15;
1789 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1803 cf =
VMUL2(cf, vq, cb_idx, sf + idx);
1811 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1824 nnz = cb_idx >> 8 & 15;
1825 sign = nnz ?
SHOW_UBITS(
re, gb, nnz) << (cb_idx >> 12) : 0;
1830 cf =
VMUL2S(cf, vq, cb_idx, sign, sf + idx);
1837 for (group = 0; group < (
AAC_SIGNE)g_len; group++, cfo+=128) {
1843 uint32_t *icf = (uint32_t *) cf;
1858 if (cb_idx == 0x0000) {
1869 for (j = 0; j < 2; j++) {
1904 unsigned v = ((
const uint32_t*)vq)[cb_idx & 15];
1905 *icf++ = (
bits & 1
U<<31) | v;
1924 if (pulse_present) {
1930 if (band_type[idx] !=
NOISE_BT && sf[idx]) {
1934 ico = co + (co > 0 ? -ico : ico);
1936 coef_base[ pulse->
pos[
i] ] = ico;
1954 const unsigned cbt_m1 = band_type[idx] - 1;
1960 for (group = 0; group < (
int)g_len; group++, cfo+=128) {
1989 k < sce->ics.swb_offset[sfb + 1];
2006 static const uint8_t gain_mode[4][3] = {
2017 uint8_t max_band =
get_bits(gb, 2);
2018 for (bd = 0; bd < max_band; bd++) {
2019 for (wd = 0; wd < gain_mode[
mode][0]; wd++) {
2020 uint8_t adjust_num =
get_bits(gb, 3);
2021 for (ad = 0; ad < adjust_num; ad++) {
2024 : gain_mode[
mode][2]));
2045 int global_gain, eld_syntax, er_syntax, pulse_present = 0;
2061 if (!common_window && !scale_flag) {
2076 if (!eld_syntax && (pulse_present =
get_bits1(gb))) {
2079 "Pulse tool not allowed in eight short sequence.\n");
2085 "Pulse data corrupt or invalid.\n");
2091 if (tns->
present && !er_syntax) {
2105 if (tns->
present && er_syntax) {
2134 int g,
i, group, idx = 0;
2142 for (group = 0; group < ics->
group_len[
g]; group++) {
2143 ac->
fdsp->butterflies_fixed(ch0 + group * 128 +
offsets[
i],
2147 for (group = 0; group < ics->
group_len[
g]; group++) {
2174 int g, group,
i, idx = 0;
2182 for (;
i < bt_run_end;
i++, idx++) {
2187 for (group = 0; group < ics->
group_len[
g]; group++)
2203 idx += bt_run_end -
i;
2219 int i,
ret, common_window, ms_present = 0;
2222 common_window = eld_syntax ||
get_bits1(gb);
2223 if (common_window) {
2234 if (ms_present == 3) {
2237 }
else if (ms_present)
2245 if (common_window) {
2259 1.09050773266525765921,
2260 1.18920711500272106672,
2304 for (
c = 0;
c < num_gain;
c++) {
2314 if ((
abs(gain_cache)-1024) >> 3 > 30)
2319 coup->
gain[
c][0] = gain_cache;
2322 for (sfb = 0; sfb < sce->
ics.
max_sfb; sfb++, idx++) {
2335 if ((
abs(gain_cache)-1024) >> 3 > 30)
2340 coup->
gain[
c][idx] = gain_cache;
2358 int num_excl_chan = 0;
2361 for (
i = 0;
i < 7;
i++)
2365 return num_excl_chan / 7;
2377 int drc_num_bands = 1;
2398 for (
i = 0;
i < drc_num_bands;
i++) {
2411 for (
i = 0;
i < drc_num_bands;
i++) {
2429 for(
i=0;
i+1<
sizeof(buf) &&
len>=8;
i++,
len-=8)
2436 if (sscanf(buf,
"libfaac %d.%d", &
major, &
minor) == 2){
2473 "SBR with 960 frame length");
2528 int bottom, top, order, start, end,
size, inc;
2550 if ((
size = end - start) <= 0)
2562 for (m = 0; m <
size; m++, start += inc)
2563 for (
i = 1;
i <=
FFMIN(m, order);
i++)
2567 for (m = 0; m <
size; m++, start += inc) {
2568 tmp[0] = coef[start];
2569 for (
i = 1;
i <=
FFMIN(m, order);
i++)
2571 for (
i = order;
i > 0;
i--)
2594 memset(in, 0, 448 *
sizeof(*in));
2601 memset(in + 1024 + 576, 0, 448 *
sizeof(*in));
2618 int16_t num_samples = 2048;
2620 if (ltp->
lag < 1024)
2621 num_samples = ltp->
lag + 1024;
2622 for (
i = 0;
i < num_samples;
i++)
2624 memset(&predTime[
i], 0, (2048 -
i) *
sizeof(*predTime));
2651 memcpy(saved_ltp, saved, 512 *
sizeof(*saved_ltp));
2652 memset(saved_ltp + 576, 0, 448 *
sizeof(*saved_ltp));
2655 for (
i = 0;
i < 64;
i++)
2658 memcpy(saved_ltp, ac->
buf_mdct + 512, 448 *
sizeof(*saved_ltp));
2659 memset(saved_ltp + 576, 0, 448 *
sizeof(*saved_ltp));
2662 for (
i = 0;
i < 64;
i++)
2667 for (
i = 0;
i < 512;
i++)
2694 for (
i = 0;
i < 1024;
i += 128)
2710 memcpy(
out, saved, 448 *
sizeof(*
out));
2718 memcpy(
out + 448 + 4*128,
temp, 64 *
sizeof(*
out));
2721 memcpy(
out + 576, buf + 64, 448 *
sizeof(*
out));
2727 memcpy( saved,
temp + 64, 64 *
sizeof(*saved));
2731 memcpy( saved + 448, buf + 7*128 + 64, 64 *
sizeof(*saved));
2733 memcpy( saved, buf + 512, 448 *
sizeof(*saved));
2734 memcpy( saved + 448, buf + 7*128 + 64, 64 *
sizeof(*saved));
2736 memcpy( saved, buf + 512, 512 *
sizeof(*saved));
2758 for (
i = 0;
i < 8;
i++)
2775 memcpy(
out, saved, 420 *
sizeof(*
out));
2783 memcpy(
out + 420 + 4*120,
temp, 60 *
sizeof(*
out));
2786 memcpy(
out + 540, buf + 60, 420 *
sizeof(*
out));
2792 memcpy( saved,
temp + 60, 60 *
sizeof(*saved));
2796 memcpy( saved + 420, buf + 7*120 + 60, 60 *
sizeof(*saved));
2798 memcpy( saved, buf + 480, 420 *
sizeof(*saved));
2799 memcpy( saved + 420, buf + 7*120 + 60, 60 *
sizeof(*saved));
2801 memcpy( saved, buf + 480, 480 *
sizeof(*saved));
2818 memcpy(
out, saved, 192 *
sizeof(*
out));
2820 memcpy(
out + 320, buf + 64, 192 *
sizeof(*
out));
2826 memcpy(saved, buf + 256, 256 *
sizeof(*saved));
2837 const int n2 = n >> 1;
2838 const int n4 = n >> 2;
2847 for (
i = 0;
i < n2;
i+=2) {
2849 temp = in[
i ]; in[
i ] = -in[n - 1 -
i]; in[n - 1 -
i] =
temp;
2850 temp = -in[
i + 1]; in[
i + 1] = in[n - 2 -
i]; in[n - 2 -
i] =
temp;
2858 for (
i = 0;
i < n;
i+=2) {
2869 for (
i = n4;
i < n2;
i ++) {
2875 for (
i = 0;
i < n2;
i ++) {
2881 for (
i = 0;
i < n4;
i ++) {
2888 memmove(saved + n, saved, 2 * n *
sizeof(*saved));
2889 memcpy( saved, buf, n *
sizeof(*saved));
2914 apply_coupling_method(ac, &cc->
ch[0], cce,
index);
2919 apply_coupling_method(ac, &cc->
ch[1], cce,
index++);
3007 int layout_map_tags,
ret;
3015 "More than one AAC RDB per ADTS frame");
3038 layout_map_tags = 2;
3039 layout_map[0][0] = layout_map[1][0] =
TYPE_SCE;
3041 layout_map[0][1] = 0;
3042 layout_map[1][1] = 1;
3089 if (chan_config < 0 || (chan_config >= 8 && chan_config < 11) || chan_config >= 13) {
3097 if (!(che=
get_che(ac, elem_type, elem_id))) {
3099 "channel element %d.%d is not allocated\n",
3100 elem_type, elem_id);
3106 switch (elem_type) {
3144 int samples = 0, multiplier, audio_found = 0, pce_found = 0;
3145 int is_dmono, sce_count = 0;
3146 int payload_alignment;
3185 if (che_presence[elem_type][elem_id]) {
3186 int error = che_presence[elem_type][elem_id] > 1;
3188 elem_type, elem_id);
3194 che_presence[elem_type][elem_id]++;
3196 if (!(che=
get_che(ac, elem_type, elem_id))) {
3198 elem_type, elem_id);
3206 switch (elem_type) {
3237 if (pce_found && !pushed) {
3250 "Not evaluating a further program_config_element as this construct is dubious at best.\n");
3270 while (elem_id > 0) {
3287 che_prev_type = elem_type;
3310 if (ac->
oc[1].
status && audio_found) {
3330 is_dmono = ac->
dmono_mode && sce_count == 2 &&
3347 int *got_frame_ptr,
AVPacket *avpkt)
3350 const uint8_t *buf = avpkt->
data;
3351 int buf_size = avpkt->
size;
3356 size_t new_extradata_size;
3359 &new_extradata_size);
3360 size_t jp_dualmono_size;
3365 if (new_extradata) {
3370 new_extradata_size * 8LL, 1);
3377 if (jp_dualmono && jp_dualmono_size > 0)
3382 if (INT_MAX / 8 <= buf_size)
3402 for (buf_offset = buf_consumed; buf_offset < buf_size; buf_offset++)
3403 if (buf[buf_offset])
3406 return buf_size > buf_offset ? buf_consumed : buf_size;
3455 #define AACDEC_FLAGS AV_OPT_FLAG_DECODING_PARAM | AV_OPT_FLAG_AUDIO_PARAM
3457 {
"dual_mono_mode",
"Select the channel to decode for dual mono",
3466 {
"channel_order",
"Order in which the channels are to be exported",
3471 {
"coded",
"order in which the channels are coded in the bitstream",
static void error(const char *err)
enum RawDataBlockType type[8]
Type of channel element to be coupled - SCE or CPE.
static void vector_pow43(int *coefs, int len)
int frame_size
Number of samples per channel in an audio frame.
CouplingPoint
The point during decoding at which channel coupling is applied.
#define FF_ENABLE_DEPRECATION_WARNINGS
@ AV_SAMPLE_FMT_FLTP
float, planar
static void skip_bits_long(GetBitContext *s, int n)
Skips the specified number of bits.
#define AV_LOG_WARNING
Something somehow does not look correct.
static av_always_inline int lcg_random(unsigned previous_val)
linear congruential pseudorandom number generator
const uint8_t ff_tns_max_bands_128[]
#define AV_EF_EXPLODE
abort decoding on minor error detection
static void apply_mid_side_stereo(AACContext *ac, ChannelElement *cpe)
Mid/Side stereo decoding; reference: 4.6.8.1.3.
static int decode_eld_specific_config(AACContext *ac, AVCodecContext *avctx, GetBitContext *gb, MPEG4AudioConfig *m4ac, int channel_config)
static void update_ltp(AACContext *ac, SingleChannelElement *sce)
Update the LTP buffer for next frame.
static int get_bits_left(GetBitContext *gb)
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
static void imdct_and_windowing(AACContext *ac, SingleChannelElement *sce)
Conduct IMDCT and windowing.
#define AV_CHANNEL_LAYOUT_STEREO
int sample_rate
samples per second
#define u(width, name, range_min, range_max)
const uint16_t ff_aac_spectral_sizes[11]
const float *const ff_aac_codebook_vector_vals[]
static int decode_fill(AACContext *ac, GetBitContext *gb, int len)
static av_cold int aac_decode_init(AVCodecContext *avctx)
static INTFLOAT aac_kbd_short_120[120]
#define INIT_VLC_STATIC(vlc, bits, a, b, c, d, e, f, g, static_size)
int skip_samples
Number of audio samples to skip at the start of the next decoded frame.
int err_recognition
Error recognition; may misdetect some more or less valid parts as errors.
#define GET_VLC(code, name, gb, table, bits, max_depth)
If the vlc code is invalid and max_depth=1, then no bits will be removed.
static int decode_prediction(AACContext *ac, IndividualChannelStream *ics, GetBitContext *gb)
void ff_cbrt_tableinit(void)
void(* vector_fmul_reverse)(float *dst, const float *src0, const float *src1, int len)
Calculate the entry wise product of two vectors of floats, and store the result in a vector of floats...
static int get_bits_count(const GetBitContext *s)
void(* subband_scale)(int *dst, int *src, int scale, int offset, int len, void *log_context)
const uint16_t *const ff_aac_codebook_vector_idx[]
This structure describes decoded (raw) audio or video data.
static int skip_data_stream_element(AACContext *ac, GetBitContext *gb)
Skip data_stream_element; reference: table 4.10.
static void apply_dependent_coupling(AACContext *ac, SingleChannelElement *target, ChannelElement *cce, int index)
Apply dependent channel coupling (applied before IMDCT).
static int AAC_RENAME() compute_lpc_coefs(const LPC_TYPE *autoc, int max_order, LPC_TYPE *lpc, int lpc_stride, int fail, int normalize)
Levinson-Durbin recursion.
const uint8_t ff_aac_num_swb_960[]
static void decode_mid_side_stereo(ChannelElement *cpe, GetBitContext *gb, int ms_present)
Decode Mid/Side data; reference: table 4.54.
@ AOT_ER_AAC_LTP
N Error Resilient Long Term Prediction.
@ AV_SAMPLE_FMT_S32P
signed 32 bits, planar
ChannelElement * tag_che_map[4][MAX_ELEM_ID]
#define FF_COMPLIANCE_STRICT
Strictly conform to all the things in the spec no matter what consequences.
const uint8_t ff_aac_num_swb_120[]
#define AV_LOG_VERBOSE
Detailed information.
int8_t used[MAX_LTP_LONG_SFB]
static int decode_band_types(AACContext *ac, enum BandType band_type[120], int band_type_run_end[120], GetBitContext *gb, IndividualChannelStream *ics)
Decode band types (section_data payload); reference: table 4.46.
static int * DEC_SQUAD(int *dst, unsigned idx)
static AVOnce aac_table_init
static void apply_prediction(AACContext *ac, SingleChannelElement *sce)
Apply AAC-Main style frequency domain prediction.
#define UPDATE_CACHE(name, gb)
enum AVChannelOrder order
Channel order used in this layout.
const uint8_t ff_aac_num_swb_480[]
static void pop_output_configuration(AACContext *ac)
Restore the previous output configuration if and only if the current configuration is unlocked.
static int assign_channels(struct elem_to_channel e2c_vec[MAX_ELEM_ID], uint8_t(*layout_map)[3], uint64_t *layout, int tags, int layer, int pos, int *current)
INTFLOAT * ret
PCM output.
#define AVERROR_UNKNOWN
Unknown error, typically from an external library.
int nb_channels
Number of channels in this layout.
static int decode_spectrum_and_dequant(AACContext *ac, INTFLOAT coef[1024], GetBitContext *gb, const INTFLOAT sf[120], int pulse_present, const Pulse *pulse, const IndividualChannelStream *ics, enum BandType band_type[120])
Decode spectral data; reference: table 4.50.
const uint16_t *const ff_swb_offset_128[]
static int init_get_bits(GetBitContext *s, const uint8_t *buffer, int bit_size)
Initialize GetBitContext.
av_cold int av_tx_init(AVTXContext **ctx, av_tx_fn *tx, enum AVTXType type, int inv, int len, const void *scale, uint64_t flags)
Initialize a transform context with the given configuration (i)MDCTs with an odd length are currently...
static int decode_pce(AVCodecContext *avctx, MPEG4AudioConfig *m4ac, uint8_t(*layout_map)[3], GetBitContext *gb, int byte_align_ref)
Decode program configuration element; reference: table 4.2.
#define FF_DEBUG_PICT_INFO
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
const uint8_t ff_tns_max_bands_1024[]
void(* butterflies_float)(float *av_restrict v1, float *av_restrict v2, int len)
Calculate the sum and difference of two vectors of floats.
static void reset_all_predictors(PredictorState *ps)
#define GET_CACHE(name, gb)
static int set_default_channel_config(AACContext *ac, AVCodecContext *avctx, uint8_t(*layout_map)[3], int *tags, int channel_config)
Set up channel positions based on a default channel configuration as specified in table 1....
static void skip_bits(GetBitContext *s, int n)
Dynamic Range Control - decoded from the bitstream but not processed further.
int num_swb
number of scalefactor window bands
static unsigned int get_bits(GetBitContext *s, int n)
Read 1-25 bits.
enum CouplingPoint coupling_point
The point during decoding at which coupling is applied.
static SDL_Window * window
const uint8_t ff_aac_num_swb_512[]
@ OC_LOCKED
Output configuration locked in place.
void(* apply_ltp)(AACContext *ac, SingleChannelElement *sce)
INTFLOAT saved[1536]
overlap
static const INTFLOAT ltp_coef[8]
INTFLOAT ret_buf[2048]
PCM output buffer.
AVChannelLayout ch_layout
Audio channel layout.
int id_select[8]
element id
#define AV_EF_BITSTREAM
detect bitstream specification deviations
static int count_paired_channels(uint8_t(*layout_map)[3], int tags, int pos, int current)
static int decode_drc_channel_exclusions(DynamicRangeControl *che_drc, GetBitContext *gb)
Parse whether channels are to be excluded from Dynamic Range Compression; reference: table 4....
int flags
AV_CODEC_FLAG_*.
#define POW_SF2_ZERO
ff_aac_pow2sf_tab index corresponding to pow(2, 0);
static void decode_gain_control(SingleChannelElement *sce, GetBitContext *gb)
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf type
static float * VMUL2(float *dst, const float *v, unsigned idx, const float *scale)
static av_always_inline float scale(float x, float s)
static void decode_ltp(LongTermPrediction *ltp, GetBitContext *gb, uint8_t max_sfb)
Decode Long Term Prediction data; reference: table 4.xx.
AVFixedDSPContext * avpriv_alloc_fixed_dsp(int bit_exact)
Allocate and initialize a fixed DSP context.
static __device__ float fabsf(float a)
uint8_t prediction_used[41]
IndividualChannelStream ics
static int aac_decode_frame(AVCodecContext *avctx, AVFrame *frame, int *got_frame_ptr, AVPacket *avpkt)
static void spectral_to_sample(AACContext *ac, int samples)
Convert spectral data to samples, applying all supported tools as appropriate.
@ AOT_ER_AAC_LC
N Error Resilient Low Complexity.
static int ff_thread_once(char *control, void(*routine)(void))
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
@ ZERO_BT
Scalefactors and spectral data are all zero.
#define FF_ARRAY_ELEMS(a)
#define FF_PROFILE_AAC_HE_V2
static av_cold int che_configure(AACContext *ac, enum ChannelPosition che_pos, int type, int id, int *channels)
Check for the channel element in the current channel position configuration.
static int decode_cpe(AACContext *ac, GetBitContext *gb, ChannelElement *cpe)
Decode a channel_pair_element; reference: table 4.4.
static int init_get_bits8(GetBitContext *s, const uint8_t *buffer, int byte_size)
Initialize GetBitContext.
#define MDCT_INIT(s, fn, len, sval)
int exclude_mask[MAX_CHANNELS]
Channels to be excluded from DRC processing.
void(* vector_pow43)(int *coefs, int len)
#define AV_CH_LAYOUT_22POINT2
#define CLOSE_READER(name, gb)
static void decode(AVCodecContext *dec_ctx, AVPacket *pkt, AVFrame *frame, FILE *outfile)
@ NOISE_BT
Spectral data are scaled white noise not coded in the bitstream.
float(* scalarproduct_float)(const float *v1, const float *v2, int len)
Calculate the scalar product of two vectors of floats.
@ AOT_ER_AAC_LD
N Error Resilient Low Delay.
static const AVClass aac_decoder_class
@ OC_TRIAL_FRAME
Output configuration under trial specified by a frame header.
INTFLOAT coeffs[1024]
coefficients for IMDCT, maybe processed
static void windowing_and_mdct_ltp(AACContext *ac, INTFLOAT *out, INTFLOAT *in, IndividualChannelStream *ics)
Apply windowing and MDCT to obtain the spectral coefficient from the predicted sample by LTP.
const uint16_t *const ff_swb_offset_960[]
static int sample_rate_idx(int rate)
static av_always_inline void reset_predict_state(PredictorState *ps)
static const int offsets[]
int num_coupled
number of target elements
static int decode_dynamic_range(DynamicRangeControl *che_drc, GetBitContext *gb)
Decode dynamic range information; reference: table 4.52.
@ AV_CHANNEL_ORDER_UNSPEC
Only the channel count is specified, without any further information about the channel order.
FF_ENABLE_DEPRECATION_WARNINGS int av_channel_layout_from_mask(AVChannelLayout *channel_layout, uint64_t mask)
Initialize a native channel layout from a bitmask indicating which channels are present.
@ OC_NONE
Output unconfigured.
@ INTENSITY_BT2
Scalefactor data are intensity stereo positions (out of phase).
static void apply_intensity_stereo(AACContext *ac, ChannelElement *cpe, int ms_present)
intensity stereo decoding; reference: 4.6.8.2.3
static VLCElem vlc_buf[16716]
#define av_assert0(cond)
assert() equivalent, that is always enabled.
static void imdct_and_windowing_eld(AACContext *ac, SingleChannelElement *sce)
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
#define SKIP_BITS(name, gb, num)
static int aac_decode_er_frame(AVCodecContext *avctx, void *data, int *got_frame_ptr, GetBitContext *gb)
static int decode_scalefactors(AACContext *ac, INTFLOAT sf[120], GetBitContext *gb, unsigned int global_gain, IndividualChannelStream *ics, enum BandType band_type[120], int band_type_run_end[120])
Decode scalefactors; reference: table 4.47.
int force_dmono_mode
0->not dmono, 1->use first channel, 2->use second channel
static int parse_adts_frame_header(AACContext *ac, GetBitContext *gb)
Individual Channel Stream.
#define SCALE_DIFF_ZERO
codebook index corresponding to zero scalefactor indices difference
INTFLOAT coef[8][4][TNS_MAX_ORDER]
#define NOISE_PRE
preamble for NOISE_BT, put in bitstream with the first noise band
const uint16_t * swb_offset
table of offsets to the lowest spectral coefficient of a scalefactor band, sfb, for a particular wind...
void ff_aac_tableinit(void)
static int decode_ga_specific_config(AACContext *ac, AVCodecContext *avctx, GetBitContext *gb, int get_bit_alignment, MPEG4AudioConfig *m4ac, int channel_config)
Decode GA "General Audio" specific configuration; reference: table 4.1.
int warned_num_aac_frames
@ INTENSITY_BT
Scalefactor data are intensity stereo positions (in phase).
#define LIBAVUTIL_VERSION_INT
Describe the class of an AVClass context structure.
static int decode_pulses(Pulse *pulse, GetBitContext *gb, const uint16_t *swb_offset, int num_swb)
Decode pulse data; reference: table 4.7.
static void flush(AVCodecContext *avctx)
#define AVERROR_PATCHWELCOME
Not yet implemented in FFmpeg, patches welcome.
enum AACOutputChannelOrder output_channel_order
void(* vector_fmul_scalar)(float *dst, const float *src, float mul, int len)
Multiply a vector of floats by a scalar float.
const float ff_aac_eld_window_480[1800]
static void imdct_and_windowing_960(AACContext *ac, SingleChannelElement *sce)
Conduct IMDCT and windowing.
const uint8_t ff_aac_num_swb_128[]
static void decode_channel_map(uint8_t layout_map[][3], enum ChannelPosition type, GetBitContext *gb, int n)
Decode an array of 4 bit element IDs, optionally interleaved with a stereo/mono switching bit.
struct AVCodecInternal * internal
Private context used for internal data.
int AAC_RENAME() ff_aac_sbr_ctx_init(AACContext *ac, SpectralBandReplication *sbr, int id_aac)
Initialize one SBR context.
const char * av_default_item_name(void *ptr)
Return the context name.
static unsigned int get_bits1(GetBitContext *s)
static av_cold int aac_decode_close(AVCodecContext *avctx)
#define LAST_SKIP_BITS(name, gb, num)
static __device__ float sqrtf(float a)
int band_top[17]
Indicates the top of the i-th DRC band in units of 4 spectral lines.
ChannelElement * che[4][MAX_ELEM_ID]
static void apply_ltp(AACContext *ac, SingleChannelElement *sce)
Apply the long term prediction.
const uint16_t *const ff_swb_offset_480[]
#define AV_CH_FRONT_CENTER
PredictorState predictor_state[MAX_PREDICTORS]
static av_always_inline int get_vlc2(GetBitContext *s, const VLCElem *table, int bits, int max_depth)
Parse a vlc code.
static int decode_audio_specific_config(AACContext *ac, AVCodecContext *avctx, MPEG4AudioConfig *m4ac, const uint8_t *data, int64_t bit_size, int sync_extension)
static int decode_cce(AACContext *ac, GetBitContext *gb, ChannelElement *che)
Decode coupling_channel_element; reference: table 4.8.
static void apply_channel_coupling(AACContext *ac, ChannelElement *cc, enum RawDataBlockType type, int elem_id, enum CouplingPoint coupling_point, void(*apply_coupling_method)(AACContext *ac, SingleChannelElement *target, ChannelElement *cce, int index))
channel coupling transformation interface
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
int band_type_run_end[120]
band type run end points
const uint8_t ff_tns_max_bands_512[]
static float * VMUL2S(float *dst, const float *v, unsigned idx, unsigned sign, const float *scale)
uint8_t layout_map[MAX_ELEM_ID *4][3]
void(* apply_tns)(INTFLOAT coef[1024], TemporalNoiseShaping *tns, IndividualChannelStream *ics, int decode)
const uint8_t ff_aac_scalefactor_bits[121]
static float * VMUL4(float *dst, const float *v, unsigned idx, const float *scale)
static float * VMUL4S(float *dst, const float *v, unsigned idx, unsigned sign, const float *scale)
const uint8_t ff_aac_pred_sfb_max[]
static const int16_t aac_channel_map[3][4][6]
static int decode_audio_specific_config_gb(AACContext *ac, AVCodecContext *avctx, MPEG4AudioConfig *m4ac, GetBitContext *gb, int get_bit_alignment, int sync_extension)
Decode audio specific configuration; reference: table 1.13.
int ff_get_buffer(AVCodecContext *avctx, AVFrame *frame, int flags)
Get a buffer for a frame.
@ AOT_ER_AAC_SCALABLE
N Error Resilient Scalable.
SingleChannelElement ch[2]
const uint16_t *const ff_swb_offset_1024[]
static void relative_align_get_bits(GetBitContext *gb, int reference_position)
@ AOT_AAC_SCALABLE
N Scalable.
An AVChannelLayout holds information about the channel layout of audio data.
void(* vector_fmul)(float *dst, const float *src0, const float *src1, int len)
Calculate the entry wise product of two vectors of floats and store the result in a vector of floats.
void ff_aac_float_common_init(void)
static INTFLOAT sine_120[120]
int warned_remapping_once
static int decode_ics(AACContext *ac, SingleChannelElement *sce, GetBitContext *gb, int common_window, int scale_flag)
Decode an individual_channel_stream payload; reference: table 4.44.
int sample_rate
Sample rate of the audio data.
static const int8_t tags_per_config[16]
static void noise_scale(int *coefs, int scale, int band_energy, int len)
static INTFLOAT sine_960[960]
enum AVSampleFormat sample_fmt
audio sample format
uint32_t ff_cbrt_tab[1<< 13]
int prog_ref_level
A reference level for the long-term program audio level for all channels combined.
void avpriv_report_missing_feature(void *avc, const char *msg,...) av_printf_format(2
Log a generic warning message about a missing feature.
const uint16_t *const ff_aac_spectral_codes[11]
OCStatus
Output configuration status.
int ff_mpeg4audio_get_config_gb(MPEG4AudioConfig *c, GetBitContext *gb, int sync_extension, void *logctx)
Parse MPEG-4 systems extradata from a potentially unaligned GetBitContext to retrieve audio configura...
static int push_output_configuration(AACContext *ac)
Save current output configuration if and only if it has been locked.
const uint8_t ff_tns_max_bands_480[]
#define OPEN_READER(name, gb)
static int assign_pair(struct elem_to_channel e2c_vec[MAX_ELEM_ID], uint8_t(*layout_map)[3], int offset, uint64_t left, uint64_t right, int pos, uint64_t *layout)
const uint16_t *const ff_swb_offset_512[]
void AAC_RENAME() ff_sbr_apply(AACContext *ac, SpectralBandReplication *sbr, int id_aac, INTFLOAT *L, INTFLOAT *R)
Apply one SBR element to one AAC element.
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf offset
static void skip_bits1(GetBitContext *s)
static int output_configure(AACContext *ac, uint8_t layout_map[MAX_ELEM_ID *4][3], int tags, enum OCStatus oc_type, int get_new_frame)
Configure output channel order based on the current program configuration element.
@ AV_CHAN_UNUSED
Channel is empty can be safely skipped.
uint8_t ms_mask[128]
Set if mid/side stereo is used for each scalefactor window band.
int dyn_rng_ctl[17]
DRC magnitude information.
av_cold void av_tx_uninit(AVTXContext **ctx)
Frees a context and sets *ctx to NULL, does nothing when *ctx == NULL.
int av_channel_layout_compare(const AVChannelLayout *chl, const AVChannelLayout *chl1)
Check whether two channel layouts are semantically the same, i.e.
#define AV_LOG_INFO
Standard information.
void ff_sine_window_init(float *window, int n)
Generate a sine window.
@ OC_GLOBAL_HDR
Output configuration set in a global header but not yet locked.
@ AOT_AAC_SSR
N (code in SoC repo) Scalable Sample Rate.
static INTFLOAT aac_kbd_long_960[960]
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel layout
@ AV_PKT_DATA_JP_DUALMONO
An AV_PKT_DATA_JP_DUALMONO side data packet indicates that the packet may contain "dual mono" audio s...
static av_cold void aac_static_table_init(void)
int nb_samples
number of audio samples (per channel) described by this frame
static void aacdec_init(AACContext *ac)
Single Channel Element - used for both SCE and LFE elements.
#define i(width, name, range_min, range_max)
static av_cold void init_sine_windows_fixed(void)
and forward the test the status of outputs and forward it to the corresponding return FFERROR_NOT_READY If the filters stores internally one or a few frame for some it can consider them to be part of the FIFO and delay acknowledging a status change accordingly Example code
static const AVOption options[]
SpectralBandReplication sbr
uint8_t * extradata
some codecs need / can use extradata like Huffman tables.
static unsigned int show_bits(GetBitContext *s, int n)
Show 1-25 bits.
uint8_t * av_packet_get_side_data(const AVPacket *pkt, enum AVPacketSideDataType type, size_t *size)
Get side information from packet.
int ff_init_vlc_sparse(VLC *vlc, int nb_bits, int nb_codes, const void *bits, int bits_wrap, int bits_size, const void *codes, int codes_wrap, int codes_size, const void *symbols, int symbols_wrap, int symbols_size, int flags)
const float ff_aac_eld_window_512[1920]
uint8_t ** extended_data
pointers to the data planes/channels.
channel element - generic struct for SCE/CPE/CCE/LFE
@ AOT_ER_AAC_ELD
N Error Resilient Enhanced Low Delay.
static void apply_independent_coupling(AACContext *ac, SingleChannelElement *target, ChannelElement *cce, int index)
Apply independent channel coupling (applied after IMDCT).
static av_always_inline void predict(PredictorState *ps, float *coef, int output_enable)
#define AV_CH_LAYOUT_NATIVE
Channel mask value used for AVCodecContext.request_channel_layout to indicate that the user requests ...
#define NOISE_PRE_BITS
length of preamble
#define FF_DEBUG_STARTCODE
static VLC vlc_spectral[11]
static av_always_inline float cbrtf(float x)
static int decode_ics_info(AACContext *ac, IndividualChannelStream *ics, GetBitContext *gb)
Decode Individual Channel Stream info; reference: table 4.6.
void av_frame_unref(AVFrame *frame)
Unreference all the buffers referenced by frame and reset the frame fields.
void * av_mallocz(size_t size)
Allocate a memory block with alignment suitable for all memory accesses (including vectors if availab...
OutputConfiguration oc[2]
static const int8_t filt[NUMTAPS *2]
static void apply_tns(INTFLOAT coef_param[1024], TemporalNoiseShaping *tns, IndividualChannelStream *ics, int decode)
Decode Temporal Noise Shaping filter coefficients and apply all-pole filters; reference: 4....
static void reset_predictor_group(PredictorState *ps, int group_num)
@ OC_TRIAL_PCE
Output configuration under trial specified by an inband PCE.
static void subband_scale(int *dst, int *src, int scale, int offset, int len, void *log_context)
DynamicRangeControl che_drc
const uint16_t *const ff_swb_offset_120[]
#define FF_PROFILE_AAC_HE
@ AOT_ER_BSAC
N Error Resilient Bit-Sliced Arithmetic Coding.
int pce_instance_tag
Indicates with which program the DRC info is associated.
#define INIT_VLC_STATIC_OVERLONG
const uint8_t ff_aac_num_swb_1024[]
#define FFSWAP(type, a, b)
const char * class_name
The name of the class; usually it is the same name as the context structure type to which the AVClass...
INTFLOAT sf[120]
scalefactors
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
const uint8_t *const ff_aac_spectral_bits[11]
int strict_std_compliance
strictly follow the standard (MPEG-4, ...).
SingleChannelElement * output_element[MAX_CHANNELS]
Points to each SingleChannelElement.
#define AACDEC_FLAGS
AVOptions for Japanese DTV specific extensions (ADTS only)
static const uint8_t * align_get_bits(GetBitContext *s)
void(* update_ltp)(AACContext *ac, SingleChannelElement *sce)
int ch_select[8]
[0] shared list of gains; [1] list of gains for right channel; [2] list of gains for left channel; [3...
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2] ... the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so ...,+,-,+,-,+,+,-,+,-,+,... hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32 - hcoeff[1] - hcoeff[2] - ... a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2} an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||......... intra?||||:Block01 :yes no||||:Block02 :....... ..........||||:Block03 ::y DC ::ref index:||||:Block04 ::cb DC ::motion x :||||......... :cr DC ::motion y :||||....... ..........|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------ ------------ ------------|||Y subbands||Cb subbands||Cr subbands||||--- ---||--- ---||--- ---|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------ ------------ ------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction ------------|\ Dequantization ------------------- \||Reference frames|\ IDWT|------- -------|Motion \|||Frame 0||Frame 1||Compensation . OBMC v -------|------- -------|--------------. \------> Frame n output Frame Frame<----------------------------------/|...|------------------- Range Coder:============Binary Range Coder:------------------- The implemented range coder is an adapted version based upon "Range encoding: an algorithm for removing redundancy from a digitised message." by G. N. N. Martin. The symbols encoded by the Snow range coder are bits(0|1). The associated probabilities are not fix but change depending on the symbol mix seen so far. bit seen|new state ---------+----------------------------------------------- 0|256 - state_transition_table[256 - old_state];1|state_transition_table[old_state];state_transition_table={ 0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:------------------------- FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1. the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
static ChannelElement * get_che(AACContext *ac, int type, int elem_id)
static void imdct_and_window(TwinVQContext *tctx, enum TwinVQFrameType ftype, int wtype, float *in, float *prev, int ch)
@ AV_CHAN_NONE
Invalid channel index.
main external API structure.
int AAC_RENAME() ff_decode_sbr_extension(AACContext *ac, SpectralBandReplication *sbr, GetBitContext *gb, int crc, int cnt, int id_aac)
Decode one SBR element.
static VLC vlc_scalefactors
#define SHOW_UBITS(name, gb, num)
int ps
-1 implicit, 1 presence
float ff_aac_pow2sf_tab[428]
void ff_aacdec_init_mips(AACContext *c)
#define NOISE_OFFSET
subtracted from global gain, used as offset for the preamble
@ AV_PKT_DATA_NEW_EXTRADATA
The AV_PKT_DATA_NEW_EXTRADATA is used to notify the codec or the format that the extradata buffer was...
static int decode_tns(AACContext *ac, TemporalNoiseShaping *tns, GetBitContext *gb, const IndividualChannelStream *ics)
Decode Temporal Noise Shaping data; reference: table 4.48.
enum WindowSequence window_sequence[2]
const uint8_t ff_mpeg4audio_channels[15]
void ff_init_ff_sine_windows(int index)
initialize the specified entry of ff_sine_windows
void av_channel_layout_uninit(AVChannelLayout *channel_layout)
Free any allocated data in the channel layout and reset the channel count to 0.
av_cold void ff_kbd_window_init(float *window, float alpha, int n)
Generate a Kaiser-Bessel Derived Window.
void AAC_RENAME() ff_aac_sbr_ctx_close(SpectralBandReplication *sbr)
Close one SBR context.
int sbr
-1 implicit, 1 presence
Filter the word “frame” indicates either a video frame or a group of audio samples
int band_incr
Number of DRC bands greater than 1 having DRC info.
#define AV_CH_FRONT_RIGHT
int av_channel_layout_copy(AVChannelLayout *dst, const AVChannelLayout *src)
Make a copy of a channel layout.
static int * DEC_UQUAD(int *dst, unsigned idx, unsigned sign)
#define FF_DISABLE_DEPRECATION_WARNINGS
AVChannelLayout ch_layout
void(* windowing_and_mdct_ltp)(AACContext *ac, INTFLOAT *out, INTFLOAT *in, IndividualChannelStream *ics)
#define AV_CODEC_FLAG_BITEXACT
Use only bitexact stuff (except (I)DCT).
static int frame_configure_elements(AVCodecContext *avctx)
static int decode_extension_payload(AACContext *ac, GetBitContext *gb, int cnt, ChannelElement *che, enum RawDataBlockType elem_type)
Decode extension data (incomplete); reference: table 4.51.
#define avpriv_request_sample(...)
void(* vector_fmul_window)(float *dst, const float *src0, const float *src1, const float *win, int len)
Overlap/add with window function.
static int count_channels(uint8_t(*layout)[3], int tags)
int dyn_rng_sgn[17]
DRC sign information; 0 - positive, 1 - negative.
This structure stores compressed data.
static void imdct_and_windowing_ld(AACContext *ac, SingleChannelElement *sce)
av_cold AVFloatDSPContext * avpriv_float_dsp_alloc(int bit_exact)
Allocate a float DSP context.
#define AVERROR_BUG
Internal bug, also see AVERROR_BUG2.
uint8_t max_sfb
number of scalefactor bands per group
INTFLOAT ltp_state[3072]
time signal for LTP
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
int interpolation_scheme
Indicates the interpolation scheme used in the SBR QMF domain.
static av_always_inline int fixed_sqrt(int x, int bits)
Calculate the square root.
void AAC_RENAME() ff_aac_sbr_init(void)
Initialize SBR.
static int aac_decode_frame_int(AVCodecContext *avctx, AVFrame *frame, int *got_frame_ptr, GetBitContext *gb, const AVPacket *avpkt)
static int * DEC_UPAIR(int *dst, unsigned idx, unsigned sign)
static int * DEC_SPAIR(int *dst, unsigned idx)
static const INTFLOAT *const tns_tmp2_map[4]
void(* imdct_and_windowing)(AACContext *ac, SingleChannelElement *sce)
enum BandType band_type[128]
band types
uint8_t use_kb_window[2]
If set, use Kaiser-Bessel window, otherwise use a sine window.
@ AOT_AAC_LC
Y Low Complexity.
@ AOT_AAC_LTP
Y Long Term Prediction.
static const float cce_scale[]
int dmono_mode
0->not dmono, 1->use first channel, 2->use second channel
int predictor_reset_group
const uint32_t ff_aac_scalefactor_code[121]
static uint64_t sniff_channel_order(uint8_t(*layout_map)[3], int tags)
static const uint8_t aac_channel_layout_map[16][16][3]
int predictor_initialized