static void search_for_pns(AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce)
{
FFPsyBand *band;
int w, g, w2, i, start, count = 0;
float *PNS = &s->scoefs[0*128], *PNS34 = &s->scoefs[1*128];
float *NOR34 = &s->scoefs[3*128];
const float lambda = s->lambda;
const float freq_mult = avctx->sample_rate/(1024.0f/sce->ics.num_windows)/2.0f;
const float thr_mult = NOISE_LAMBDA_REPLACE*(100.0f/lambda);
const float spread_threshold = NOISE_SPREAD_THRESHOLD*(lambda/100.f);
for (w = 0; w < sce->ics.num_windows; w += sce->ics.group_len[w]) {
start = 0;
for (g = 0; g < sce->ics.num_swb; g++) {
int noise_sfi, try_pns = 0;
float dist1 = 0.0f, dist2 = 0.0f, noise_amp;
float energy = 0.0f, threshold = 0.0f, spread = 0.0f;
if (start*freq_mult < NOISE_LOW_LIMIT) {
start += sce->ics.swb_sizes[g];
continue;
}
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
energy += band->energy;
spread += band->spread;
threshold += band->threshold;
}
sce->pns_ener[w*16+g] = energy;
if (sce->zeroes[w*16+g]) {
try_pns = 1;
} else if (energy < threshold) {
try_pns = 1;
} else if (spread > spread_threshold) {
try_pns = 0;
} else if (energy < threshold*thr_mult) {
try_pns = 1;
}
if (!try_pns || !energy) {
start += sce->ics.swb_sizes[g];
continue;
}
noise_sfi = av_clip(roundf(log2f(energy)*2), -100, 155); /* Quantize */
noise_amp = -ff_aac_pow2sf_tab[noise_sfi + POW_SF2_ZERO]; /* Dequantize */
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
float band_energy, scale;
band = &s->psy.ch[s->cur_channel+0].psy_bands[(w+w2)*16+g];
for (i = 0; i < sce->ics.swb_sizes[g]; i++)
PNS[i] = s->random_state = lcg_random(s->random_state);
band_energy = s->fdsp->scalarproduct_float(PNS, PNS, sce->ics.swb_sizes[g]);
scale = noise_amp/sqrtf(band_energy);
s->fdsp->vector_fmul_scalar(PNS, PNS, scale, sce->ics.swb_sizes[g]);
abs_pow34_v(NOR34, &sce->coeffs[start+(w+w2)*128], sce->ics.swb_sizes[g]);
abs_pow34_v(PNS34, PNS, sce->ics.swb_sizes[g]);
dist1 += quantize_band_cost(s, &sce->coeffs[start + (w+w2)*128],
NOR34,
sce->ics.swb_sizes[g],
sce->sf_idx[(w+w2)*16+g],
sce->band_alt[(w+w2)*16+g],
lambda/band->threshold, INFINITY, NULL, 0);
dist2 += quantize_band_cost(s, PNS,
PNS34,
sce->ics.swb_sizes[g],
noise_sfi,
NOISE_BT,
lambda/band->threshold, INFINITY, NULL, 0);
}
if (dist2 < dist1) {
sce->band_type[w*16+g] = NOISE_BT;
sce->zeroes[w*16+g] = 0;
if (sce->band_type[w*16+g-1] != NOISE_BT && /* Prevent holes */
sce->band_type[w*16+g-2] == NOISE_BT) {
sce->band_type[w*16+g-1] = NOISE_BT;
sce->zeroes[w*16+g-1] = 0;
}
count++;
}
start += sce->ics.swb_sizes[g];
}
}
}
static void search_for_pns(AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce)
{
FFPsyBand *band;
int w, g, w2, i;
int wlen = 1024 / sce->ics.num_windows;
int bandwidth, cutoff;
float *PNS = &s->scoefs[0*128], *PNS34 = &s->scoefs[1*128];
float *NOR34 = &s->scoefs[3*128];
const float lambda = s->lambda;
const float freq_mult = avctx->sample_rate*0.5f/wlen;
const float thr_mult = NOISE_LAMBDA_REPLACE*(100.0f/lambda);
const float spread_threshold = FFMIN(0.75f, NOISE_SPREAD_THRESHOLD*FFMAX(0.5f, lambda/100.f));
const float dist_bias = av_clipf(4.f * 120 / lambda, 0.25f, 4.0f);
const float pns_transient_energy_r = FFMIN(0.7f, lambda / 140.f);
int refbits = avctx->bit_rate * 1024.0 / avctx->sample_rate
/ ((avctx->flags & CODEC_FLAG_QSCALE) ? 2.0f : avctx->channels)
* (lambda / 120.f);
/** Keep this in sync with twoloop's cutoff selection */
float rate_bandwidth_multiplier = 1.5f;
int frame_bit_rate = (avctx->flags & CODEC_FLAG_QSCALE)
? (refbits * rate_bandwidth_multiplier * avctx->sample_rate / 1024)
: (avctx->bit_rate / avctx->channels);
frame_bit_rate *= 1.15f;
if (avctx->cutoff > 0) {
bandwidth = avctx->cutoff;
} else {
bandwidth = FFMAX(3000, AAC_CUTOFF_FROM_BITRATE(frame_bit_rate, 1, avctx->sample_rate));
}
cutoff = bandwidth * 2 * wlen / avctx->sample_rate;
memcpy(sce->band_alt, sce->band_type, sizeof(sce->band_type));
for (w = 0; w < sce->ics.num_windows; w += sce->ics.group_len[w]) {
int wstart = w*128;
for (g = 0; g < sce->ics.num_swb; g++) {
int noise_sfi;
float dist1 = 0.0f, dist2 = 0.0f, noise_amp;
float pns_energy = 0.0f, pns_tgt_energy, energy_ratio, dist_thresh;
float sfb_energy = 0.0f, threshold = 0.0f, spread = 2.0f;
float min_energy = -1.0f, max_energy = 0.0f;
const int start = wstart+sce->ics.swb_offset[g];
const float freq = (start-wstart)*freq_mult;
const float freq_boost = FFMAX(0.88f*freq/NOISE_LOW_LIMIT, 1.0f);
if (freq < NOISE_LOW_LIMIT || (start-wstart) >= cutoff)
continue;
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
sfb_energy += band->energy;
spread = FFMIN(spread, band->spread);
threshold += band->threshold;
if (!w2) {
min_energy = max_energy = band->energy;
} else {
min_energy = FFMIN(min_energy, band->energy);
max_energy = FFMAX(max_energy, band->energy);
}
}
/* Ramps down at ~8000Hz and loosens the dist threshold */
dist_thresh = av_clipf(2.5f*NOISE_LOW_LIMIT/freq, 0.5f, 2.5f) * dist_bias;
/* PNS is acceptable when all of these are true:
* 1. high spread energy (noise-like band)
* 2. near-threshold energy (high PE means the random nature of PNS content will be noticed)
* 3. on short window groups, all windows have similar energy (variations in energy would be destroyed by PNS)
*
* At this stage, point 2 is relaxed for zeroed bands near the noise threshold (hole avoidance is more important)
*/
if (((sce->zeroes[w*16+g] || !sce->band_alt[w*16+g]) && sfb_energy < threshold*sqrtf(1.5f/freq_boost)) || spread < spread_threshold ||
(!sce->zeroes[w*16+g] && sce->band_alt[w*16+g] && sfb_energy > threshold*thr_mult*freq_boost) ||
min_energy < pns_transient_energy_r * max_energy ) {
sce->pns_ener[w*16+g] = sfb_energy;
continue;
}
pns_tgt_energy = sfb_energy*FFMIN(1.0f, spread*spread);
noise_sfi = av_clip(roundf(log2f(pns_tgt_energy)*2), -100, 155); /* Quantize */
noise_amp = -ff_aac_pow2sf_tab[noise_sfi + POW_SF2_ZERO]; /* Dequantize */
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
float band_energy, scale, pns_senergy;
const int start_c = (w+w2)*128+sce->ics.swb_offset[g];
band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
for (i = 0; i < sce->ics.swb_sizes[g]; i++)
PNS[i] = s->random_state = lcg_random(s->random_state);
band_energy = s->fdsp->scalarproduct_float(PNS, PNS, sce->ics.swb_sizes[g]);
scale = noise_amp/sqrtf(band_energy);
s->fdsp->vector_fmul_scalar(PNS, PNS, scale, sce->ics.swb_sizes[g]);
pns_senergy = s->fdsp->scalarproduct_float(PNS, PNS, sce->ics.swb_sizes[g]);
pns_energy += pns_senergy;
abs_pow34_v(NOR34, &sce->coeffs[start_c], sce->ics.swb_sizes[g]);
abs_pow34_v(PNS34, PNS, sce->ics.swb_sizes[g]);
dist1 += quantize_band_cost(s, &sce->coeffs[start_c],
NOR34,
sce->ics.swb_sizes[g],
sce->sf_idx[(w+w2)*16+g],
sce->band_alt[(w+w2)*16+g],
lambda/band->threshold, INFINITY, NULL, NULL, 0);
/* Estimate rd on average as 5 bits for SF, 4 for the CB, plus spread energy * lambda/thr */
dist2 += band->energy/(band->spread*band->spread)*lambda*dist_thresh/band->threshold;
}
if (g && sce->sf_idx[(w+w2)*16+g-1] == NOISE_BT) {
dist2 += 5;
} else {
dist2 += 9;
}
energy_ratio = pns_tgt_energy/pns_energy; /* Compensates for quantization error */
sce->pns_ener[w*16+g] = energy_ratio*pns_tgt_energy;
if (sce->zeroes[w*16+g] || !sce->band_alt[w*16+g] || (energy_ratio > 0.85f && energy_ratio < 1.25f && dist2 < dist1)) {
sce->band_type[w*16+g] = NOISE_BT;
sce->zeroes[w*16+g] = 0;
}
}
}
}
static void search_for_pns(AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce)
{
FFPsyBand *band;
int w, g, w2, i;
float *PNS = &s->scoefs[0*128], *PNS34 = &s->scoefs[1*128];
float *NOR34 = &s->scoefs[3*128];
const float lambda = s->lambda;
const float freq_mult = avctx->sample_rate/(1024.0f/sce->ics.num_windows)/2.0f;
const float thr_mult = NOISE_LAMBDA_REPLACE*(100.0f/lambda);
const float spread_threshold = NOISE_SPREAD_THRESHOLD*(lambda/100.f);
if (sce->ics.window_sequence[0] == EIGHT_SHORT_SEQUENCE)
return;
for (w = 0; w < sce->ics.num_windows; w += sce->ics.group_len[w]) {
for (g = 0; g < sce->ics.num_swb; g++) {
int noise_sfi;
float dist1 = 0.0f, dist2 = 0.0f, noise_amp;
float pns_energy = 0.0f, energy_ratio, dist_thresh;
float sfb_energy = 0.0f, threshold = 0.0f, spread = 0.0f;
const int start = sce->ics.swb_offset[w*16+g];
const float freq = start*freq_mult;
const float freq_boost = FFMAX(0.88f*freq/NOISE_LOW_LIMIT, 1.0f);
if (freq < NOISE_LOW_LIMIT || avctx->cutoff && freq >= avctx->cutoff)
continue;
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
sfb_energy += band->energy;
spread += band->spread;
threshold += band->threshold;
}
/* Ramps down at ~8000Hz and loosens the dist threshold */
dist_thresh = FFMIN(2.5f*NOISE_LOW_LIMIT/freq, 1.27f);
if (sce->zeroes[w*16+g] || spread < spread_threshold ||
sfb_energy > threshold*thr_mult*freq_boost) {
sce->pns_ener[w*16+g] = sfb_energy;
continue;
}
noise_sfi = av_clip(roundf(log2f(sfb_energy)*2), -100, 155); /* Quantize */
noise_amp = -ff_aac_pow2sf_tab[noise_sfi + POW_SF2_ZERO]; /* Dequantize */
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
float band_energy, scale;
const int start_c = sce->ics.swb_offset[(w+w2)*16+g];
band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
for (i = 0; i < sce->ics.swb_sizes[g]; i++)
PNS[i] = s->random_state = lcg_random(s->random_state);
band_energy = s->fdsp->scalarproduct_float(PNS, PNS, sce->ics.swb_sizes[g]);
scale = noise_amp/sqrtf(band_energy);
s->fdsp->vector_fmul_scalar(PNS, PNS, scale, sce->ics.swb_sizes[g]);
pns_energy += s->fdsp->scalarproduct_float(PNS, PNS, sce->ics.swb_sizes[g]);
abs_pow34_v(NOR34, &sce->coeffs[start_c], sce->ics.swb_sizes[g]);
abs_pow34_v(PNS34, PNS, sce->ics.swb_sizes[g]);
dist1 += quantize_band_cost(s, &sce->coeffs[start_c],
NOR34,
sce->ics.swb_sizes[g],
sce->sf_idx[(w+w2)*16+g],
sce->band_alt[(w+w2)*16+g],
lambda/band->threshold, INFINITY, NULL, 0);
dist2 += quantize_band_cost(s, PNS,
PNS34,
sce->ics.swb_sizes[g],
noise_sfi,
NOISE_BT,
lambda/band->threshold, INFINITY, NULL, 0);
}
energy_ratio = sfb_energy/pns_energy; /* Compensates for quantization error */
sce->pns_ener[w*16+g] = energy_ratio*sfb_energy;
if (energy_ratio > 0.85f && energy_ratio < 1.25f && dist1/dist2 > dist_thresh) {
sce->band_type[w*16+g] = NOISE_BT;
sce->zeroes[w*16+g] = 0;
if (sce->band_type[w*16+g-1] != NOISE_BT && /* Prevent holes */
sce->band_type[w*16+g-2] == NOISE_BT) {
sce->band_type[w*16+g-1] = NOISE_BT;
sce->zeroes[w*16+g-1] = 0;
}
}
}
}
}
