Avoid clipping due to quantization noise to produce audible
artifacts, by detecting near-clipping signals and both attenuating
them a little and encoding escape-encoded bands (usually the
loudest) rounding towards zero instead of nearest, which tends to
decrease overall energy and thus clipping.
Currently fate tests measure numerical error so this change makes
tests using asynth (which are near clipping) report higher error
not less, because of window attenuation. Yet, they sound better,
not worse (albeit subtle, other samples aren't subtle at all).
Only measuring psychoacoustically weighted error would make for
a representative test, so that will be left for a future patch.
From 9da94f02574b34025a56c225c11269802f49949b Mon Sep 17 00:00:00 2001
From: Claudio Freire <klaussfre...@gmail.com>
Date: Fri, 17 Jul 2015 05:47:25 -0300
Subject: [PATCH] AAC Encoder: clipping avoidance
Avoid clipping due to quantization noise to produce audible
artifacts, by detecting near-clipping signals and both attenuating
them a little and encoding escape-encoded bands (usually the
loudest) rounding towards zero instead of nearest, which tends to
decrease overall energy and thus clipping.
Currently fate tests measure numerical error so this change makes
tests using asynth (which are near clipping) report higher error
not less, because of window attenuation. Yet, they sound better,
not worse (albeit subtle, other samples aren't subtle at all).
Only measuring psychoacoustically weighted error would make for
a representative test, so that will be left for a future patch.
---
libavcodec/aac.h | 1 +
libavcodec/aaccoder.c | 108 ++++++++++++++++++++++++++++++++------------------
libavcodec/aacenc.c | 28 ++++++++++++-
libavcodec/aacenc.h | 2 +-
libavcodec/aacpsy.c | 31 +++++++++++++++
libavcodec/psymodel.h | 1 +
tests/fate/aac.mak | 2 +-
7 files changed, 132 insertions(+), 41 deletions(-)
diff --git a/libavcodec/aac.h b/libavcodec/aac.h
index f6fd446..7548897 100644
--- a/libavcodec/aac.h
+++ b/libavcodec/aac.h
@@ -230,6 +230,7 @@ typedef struct IndividualChannelStream {
int predictor_initialized;
int predictor_reset_group;
uint8_t prediction_used[41];
+ uint8_t window_clipping[8]; ///< set if a certain window is near clipping
} IndividualChannelStream;
/**
diff --git a/libavcodec/aaccoder.c b/libavcodec/aaccoder.c
index 5bdba46..d606119 100644
--- a/libavcodec/aaccoder.c
+++ b/libavcodec/aaccoder.c
@@ -78,6 +78,9 @@ static const uint8_t * const run_value_bits[2] = {
run_value_bits_long, run_value_bits_short
};
+#define ROUND_STANDARD 0.4054f
+#define ROUND_TO_ZERO 0.1054f
+
/** Map to convert values from BandCodingPath index to a codebook index **/
static const uint8_t aac_cb_out_map[CB_TOT_ALL] = {0,1,2,3,4,5,6,7,8,9,10,11,13,14,15};
/** Inverse map to convert from codebooks to BandCodingPath indices **/
@@ -88,20 +91,20 @@ static const uint8_t aac_cb_in_map[CB_TOT_ALL+1] = {0,1,2,3,4,5,6,7,8,9,10,11,0,
* @return absolute value of the quantized coefficient
* @see 3GPP TS26.403 5.6.2 "Scalefactor determination"
*/
-static av_always_inline int quant(float coef, const float Q)
+static av_always_inline int quant(float coef, const float Q, const float rounding)
{
float a = coef * Q;
- return sqrtf(a * sqrtf(a)) + 0.4054;
+ return sqrtf(a * sqrtf(a)) + rounding;
}
static void quantize_bands(int *out, const float *in, const float *scaled,
- int size, float Q34, int is_signed, int maxval)
+ int size, float Q34, int is_signed, int maxval, const float rounding)
{
int i;
double qc;
for (i = 0; i < size; i++) {
qc = scaled[i] * Q34;
- out[i] = (int)FFMIN(qc + 0.4054, (double)maxval);
+ out[i] = (int)FFMIN(qc + rounding, (double)maxval);
if (is_signed && in[i] < 0.0f) {
out[i] = -out[i];
}
@@ -133,7 +136,8 @@ static av_always_inline float quantize_and_encode_band_cost_template(
const float *scaled, int size, int scale_idx,
int cb, const float lambda, const float uplim,
int *bits, int BT_ZERO, int BT_UNSIGNED,
- int BT_PAIR, int BT_ESC, int BT_NOISE, int BT_STEREO)
+ int BT_PAIR, int BT_ESC, int BT_NOISE, int BT_STEREO,
+ const float ROUNDING)
{
const int q_idx = POW_SF2_ZERO - scale_idx + SCALE_ONE_POS - SCALE_DIV_512;
const float Q = ff_aac_pow2sf_tab [q_idx];
@@ -157,7 +161,7 @@ static av_always_inline float quantize_and_encode_band_cost_template(
abs_pow34_v(s->scoefs, in, size);
scaled = s->scoefs;
}
- quantize_bands(s->qcoefs, in, scaled, size, Q34, !BT_UNSIGNED, aac_cb_maxval[cb]);
+ quantize_bands(s->qcoefs, in, scaled, size, Q34, !BT_UNSIGNED, aac_cb_maxval[cb], ROUNDING);
if (BT_UNSIGNED) {
off = 0;
} else {
@@ -184,7 +188,7 @@ static av_always_inline float quantize_and_encode_band_cost_template(
di = t - CLIPPED_ESCAPE;
curbits += 21;
} else {
- int c = av_clip_uintp2(quant(t, Q), 13);
+ int c = av_clip_uintp2(quant(t, Q, ROUNDING), 13);
di = t - c*cbrtf(c)*IQ;
curbits += av_log2(c)*2 - 4 + 1;
}
@@ -214,7 +218,7 @@ static av_always_inline float quantize_and_encode_band_cost_template(
if (BT_ESC) {
for (j = 0; j < 2; j++) {
if (ff_aac_codebook_vectors[cb-1][curidx*2+j] == 64.0f) {
- int coef = av_clip_uintp2(quant(fabsf(in[i+j]), Q), 13);
+ int coef = av_clip_uintp2(quant(fabsf(in[i+j]), Q, ROUNDING), 13);
int len = av_log2(coef);
put_bits(pb, len - 4 + 1, (1 << (len - 4 + 1)) - 2);
@@ -239,7 +243,7 @@ static float quantize_and_encode_band_cost_NONE(struct AACEncContext *s, PutBitC
return 0.0f;
}
-#define QUANTIZE_AND_ENCODE_BAND_COST_FUNC(NAME, BT_ZERO, BT_UNSIGNED, BT_PAIR, BT_ESC, BT_NOISE, BT_STEREO) \
+#define QUANTIZE_AND_ENCODE_BAND_COST_FUNC(NAME, BT_ZERO, BT_UNSIGNED, BT_PAIR, BT_ESC, BT_NOISE, BT_STEREO, ROUNDING) \
static float quantize_and_encode_band_cost_ ## NAME( \
struct AACEncContext *s, \
PutBitContext *pb, const float *in, \
@@ -249,17 +253,19 @@ static float quantize_and_encode_band_cost_ ## NAME(
return quantize_and_encode_band_cost_template( \
s, pb, in, scaled, size, scale_idx, \
BT_ESC ? ESC_BT : cb, lambda, uplim, bits, \
- BT_ZERO, BT_UNSIGNED, BT_PAIR, BT_ESC, BT_NOISE, BT_STEREO); \
+ BT_ZERO, BT_UNSIGNED, BT_PAIR, BT_ESC, BT_NOISE, BT_STEREO, \
+ ROUNDING); \
}
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(ZERO, 1, 0, 0, 0, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(SQUAD, 0, 0, 0, 0, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(UQUAD, 0, 1, 0, 0, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(SPAIR, 0, 0, 1, 0, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(UPAIR, 0, 1, 1, 0, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(ESC, 0, 1, 1, 1, 0, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(NOISE, 0, 0, 0, 0, 1, 0)
-QUANTIZE_AND_ENCODE_BAND_COST_FUNC(STEREO,0, 0, 0, 0, 0, 1)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(ZERO, 1, 0, 0, 0, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(SQUAD, 0, 0, 0, 0, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(UQUAD, 0, 1, 0, 0, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(SPAIR, 0, 0, 1, 0, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(UPAIR, 0, 1, 1, 0, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(ESC, 0, 1, 1, 1, 0, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(ESC_RTZ, 0, 1, 1, 1, 0, 0, ROUND_TO_ZERO)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(NOISE, 0, 0, 0, 0, 1, 0, ROUND_STANDARD)
+QUANTIZE_AND_ENCODE_BAND_COST_FUNC(STEREO,0, 0, 0, 0, 0, 1, ROUND_STANDARD)
static float (*const quantize_and_encode_band_cost_arr[])(
struct AACEncContext *s,
@@ -285,28 +291,52 @@ static float (*const quantize_and_encode_band_cost_arr[])(
quantize_and_encode_band_cost_STEREO,
};
+static float (*const quantize_and_encode_band_cost_rtz_arr[])(
+ struct AACEncContext *s,
+ PutBitContext *pb, const float *in,
+ const float *scaled, int size, int scale_idx,
+ int cb, const float lambda, const float uplim,
+ int *bits) = {
+ quantize_and_encode_band_cost_ZERO,
+ quantize_and_encode_band_cost_SQUAD,
+ quantize_and_encode_band_cost_SQUAD,
+ quantize_and_encode_band_cost_UQUAD,
+ quantize_and_encode_band_cost_UQUAD,
+ quantize_and_encode_band_cost_SPAIR,
+ quantize_and_encode_band_cost_SPAIR,
+ quantize_and_encode_band_cost_UPAIR,
+ quantize_and_encode_band_cost_UPAIR,
+ quantize_and_encode_band_cost_UPAIR,
+ quantize_and_encode_band_cost_UPAIR,
+ quantize_and_encode_band_cost_ESC_RTZ,
+ quantize_and_encode_band_cost_NONE, /* CB 12 doesn't exist */
+ quantize_and_encode_band_cost_NOISE,
+ quantize_and_encode_band_cost_STEREO,
+ quantize_and_encode_band_cost_STEREO,
+};
+
#define quantize_and_encode_band_cost( \
s, pb, in, scaled, size, scale_idx, cb, \
- lambda, uplim, bits) \
- quantize_and_encode_band_cost_arr[cb]( \
+ lambda, uplim, bits, rtz) \
+ ((rtz) ? quantize_and_encode_band_cost_rtz_arr : quantize_and_encode_band_cost_arr)[cb]( \
s, pb, in, scaled, size, scale_idx, cb, \
lambda, uplim, bits)
static float quantize_band_cost(struct AACEncContext *s, const float *in,
const float *scaled, int size, int scale_idx,
int cb, const float lambda, const float uplim,
- int *bits)
+ int *bits, int rtz)
{
return quantize_and_encode_band_cost(s, NULL, in, scaled, size, scale_idx,
- cb, lambda, uplim, bits);
+ cb, lambda, uplim, bits, rtz);
}
static void quantize_and_encode_band(struct AACEncContext *s, PutBitContext *pb,
const float *in, int size, int scale_idx,
- int cb, const float lambda)
+ int cb, const float lambda, int rtz)
{
quantize_and_encode_band_cost(s, pb, in, NULL, size, scale_idx, cb, lambda,
- INFINITY, NULL);
+ INFINITY, NULL, rtz);
}
static float find_max_val(int group_len, int swb_size, const float *scaled) {
@@ -396,7 +426,7 @@ static void encode_window_bands_info(AACEncContext *s, SingleChannelElement *sce
rd += quantize_band_cost(s, sce->coeffs + start + w*128,
s->scoefs + start + w*128, size,
sce->sf_idx[(win+w)*16+swb], aac_cb_out_map[cb],
- lambda / band->threshold, INFINITY, NULL);
+ lambda / band->threshold, INFINITY, NULL, 0);
}
cost_stay_here = path[swb][cb].cost + rd;
cost_get_here = minrd + rd + run_bits + 4;
@@ -526,9 +556,9 @@ static void codebook_trellis_rate(AACEncContext *s, SingleChannelElement *sce,
for (w = 0; w < group_len; w++) {
bits += quantize_band_cost(s, sce->coeffs + start + w*128,
s->scoefs + start + w*128, size,
- sce->sf_idx[(win+w)*16+swb],
+ sce->sf_idx[win*16+swb],
aac_cb_out_map[cb],
- 0, INFINITY, NULL);
+ 0, INFINITY, NULL, 0);
}
cost_stay_here = path[swb][cb].cost + bits;
cost_get_here = minbits + bits + run_bits + 4;
@@ -748,7 +778,7 @@ static void search_for_quantizers_anmr(AVCodecContext *avctx, AACEncContext *s,
for (w2 = 0; w2 < sce->ics.group_len[w]; w2++) {
FFPsyBand *band = &s->psy.ch[s->cur_channel].psy_bands[(w+w2)*16+g];
dist += quantize_band_cost(s, coefs + w2*128, s->scoefs + start + w2*128, sce->ics.swb_sizes[g],
- q + q0, cb, lambda / band->threshold, INFINITY, NULL);
+ q + q0, cb, lambda / band->threshold, INFINITY, NULL, 0);
}
minrd = FFMIN(minrd, dist);
@@ -894,7 +924,8 @@ static void search_for_quantizers_twoloop(AVCodecContext *avctx,
cb,
1.0f,
INFINITY,
- &b);
+ &b,
+ 0);
bits += b;
}
dists[w*16+g] = dist - bits;
@@ -1060,11 +1091,12 @@ static void search_for_quantizers_faac(AVCodecContext *avctx, AACEncContext *s,
ESC_BT,
lambda,
INFINITY,
- &b);
+ &b,
+ 0);
dist -= b;
}
dist *= 1.0f / 512.0f / lambda;
- quant_max = quant(maxq[w*16+g], ff_aac_pow2sf_tab[POW_SF2_ZERO - scf + SCALE_ONE_POS - SCALE_DIV_512]);
+ quant_max = quant(maxq[w*16+g], ff_aac_pow2sf_tab[POW_SF2_ZERO - scf + SCALE_ONE_POS - SCALE_DIV_512], ROUND_STANDARD);
if (quant_max >= 8191) { // too much, return to the previous quantizer
sce->sf_idx[w*16+g] = prev_scf;
break;
@@ -1241,19 +1273,19 @@ static void search_for_is(AACEncContext *s, AVCodecContext *avctx, ChannelElemen
sce0->ics.swb_sizes[g],
sce0->sf_idx[(w+w2)*16+g],
sce0->band_type[(w+w2)*16+g],
- lambda / band0->threshold, INFINITY, NULL);
+ lambda / band0->threshold, INFINITY, NULL, 0);
dist1 += quantize_band_cost(s, sce1->coeffs + start + (w+w2)*128,
R34,
sce1->ics.swb_sizes[g],
sce1->sf_idx[(w+w2)*16+g],
sce1->band_type[(w+w2)*16+g],
- lambda / band1->threshold, INFINITY, NULL);
+ lambda / band1->threshold, INFINITY, NULL, 0);
dist2 += quantize_band_cost(s, IS,
I34,
sce0->ics.swb_sizes[g],
is_sf_idx,
is_band_type,
- lambda / minthr, INFINITY, NULL);
+ lambda / minthr, INFINITY, NULL, 0);
for (i = 0; i < sce0->ics.swb_sizes[g]; i++) {
dist_spec_err += (L34[i] - I34[i])*(L34[i] - I34[i]);
dist_spec_err += (R34[i] - I34[i]*e01_34)*(R34[i] - I34[i]*e01_34);
@@ -1314,25 +1346,25 @@ static void search_for_ms(AACEncContext *s, ChannelElement *cpe,
sce0->ics.swb_sizes[g],
sce0->sf_idx[(w+w2)*16+g],
sce0->band_type[(w+w2)*16+g],
- lambda / band0->threshold, INFINITY, NULL);
+ lambda / band0->threshold, INFINITY, NULL, 0);
dist1 += quantize_band_cost(s, sce1->coeffs + start + (w+w2)*128,
R34,
sce1->ics.swb_sizes[g],
sce1->sf_idx[(w+w2)*16+g],
sce1->band_type[(w+w2)*16+g],
- lambda / band1->threshold, INFINITY, NULL);
+ lambda / band1->threshold, INFINITY, NULL, 0);
dist2 += quantize_band_cost(s, M,
M34,
sce0->ics.swb_sizes[g],
sce0->sf_idx[(w+w2)*16+g],
sce0->band_type[(w+w2)*16+g],
- lambda / maxthr, INFINITY, NULL);
+ lambda / maxthr, INFINITY, NULL, 0);
dist2 += quantize_band_cost(s, S,
S34,
sce1->ics.swb_sizes[g],
sce1->sf_idx[(w+w2)*16+g],
sce1->band_type[(w+w2)*16+g],
- lambda / minthr, INFINITY, NULL);
+ lambda / minthr, INFINITY, NULL, 0);
}
cpe->ms_mask[w*16+g] = dist2 < dist1;
}
diff --git a/libavcodec/aacenc.c b/libavcodec/aacenc.c
index f05f51b..6ff95b1 100644
--- a/libavcodec/aacenc.c
+++ b/libavcodec/aacenc.c
@@ -46,6 +46,7 @@
#include "psymodel.h"
#define AAC_MAX_CHANNELS 6
+#define CLIP_AVOIDANCE_FACTOR 0.95f
#define ERROR_IF(cond, ...) \
if (cond) { \
@@ -473,7 +474,29 @@ static void encode_spectral_coeffs(AACEncContext *s, SingleChannelElement *sce)
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, w2;
+
+ for (w = 0; w < sce->ics.num_windows; w += sce->ics.group_len[w]) {
+ start = 0;
+ for (i = 0; i < sce->ics.max_sfb; i++) {
+ if (sce->ics.window_clipping[w]) {
+ for (w2 = w; w2 < w + sce->ics.group_len[w]; w2++) {
+ float *swb_coeffs = sce->coeffs + start + w2*128;
+ for (j = 0; j < sce->ics.swb_sizes[i]; j++)
+ swb_coeffs[j] *= CLIP_AVOIDANCE_FACTOR;
+ }
+ }
start += sce->ics.swb_sizes[i];
}
}
@@ -608,12 +631,15 @@ static int aac_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
ics->num_swb = tag == TYPE_LFE ? ics->num_swb : s->psy.num_bands[ics->num_windows == 8];
for (w = 0; w < ics->num_windows; w++)
ics->group_len[w] = wi[ch].grouping[w];
+ for (w = 0; w < ics->num_windows; w++)
+ ics->window_clipping[w] = wi[ch].clipping[w];
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;
}
diff --git a/libavcodec/aacenc.h b/libavcodec/aacenc.h
index 966c708..f1134fa 100644
--- a/libavcodec/aacenc.h
+++ b/libavcodec/aacenc.h
@@ -54,7 +54,7 @@ typedef struct AACCoefficientsEncoder {
void (*encode_window_bands_info)(struct AACEncContext *s, SingleChannelElement *sce,
int win, int group_len, const float lambda);
void (*quantize_and_encode_band)(struct AACEncContext *s, PutBitContext *pb, const float *in, int size,
- int scale_idx, int cb, const float lambda);
+ int scale_idx, int cb, const float lambda, int rtz);
void (*set_special_band_scalefactors)(struct AACEncContext *s, SingleChannelElement *sce);
void (*search_for_pns)(struct AACEncContext *s, AVCodecContext *avctx, SingleChannelElement *sce, const float lambda);
void (*search_for_ms)(struct AACEncContext *s, ChannelElement *cpe, const float lambda);
diff --git a/libavcodec/aacpsy.c b/libavcodec/aacpsy.c
index b16f6b9..fbd545d 100644
--- a/libavcodec/aacpsy.c
+++ b/libavcodec/aacpsy.c
@@ -95,6 +95,8 @@ enum {
#define AAC_NUM_BLOCKS_SHORT 8 ///< number of blocks in a short sequence
#define PSY_LAME_NUM_SUBBLOCKS 3 ///< Number of sub-blocks in each short block
+#define CLIPPING_THRESHOLD 0.98f
+
/**
* @}
*/
@@ -837,6 +839,7 @@ static FFPsyWindowInfo psy_lame_window(FFPsyContext *ctx, const float *audio,
int grouping = 0;
int uselongblock = 1;
int attacks[AAC_NUM_BLOCKS_SHORT + 1] = { 0 };
+ uint8_t clippings[AAC_NUM_BLOCKS_SHORT];
int i;
FFPsyWindowInfo wi = { { 0 } };
@@ -926,14 +929,35 @@ static FFPsyWindowInfo psy_lame_window(FFPsyContext *ctx, const float *audio,
lame_apply_block_type(pch, &wi, uselongblock);
+ /* Calculate input sample maximums and evaluate clipping risk */
+ if (audio) {
+ for (i = 0; i < AAC_NUM_BLOCKS_SHORT; i++) {
+ const float *wbuf = audio + i * AAC_BLOCK_SIZE_SHORT;
+ float max = 0;
+ int j;
+ for (j = 0; j < AAC_BLOCK_SIZE_SHORT; j++)
+ max = FFMAX(max, fabsf(wbuf[j]));
+ clippings[i] = (max > CLIPPING_THRESHOLD) ? 1 : 0;
+ }
+ } else {
+ for (i = 0; i < 8; i++)
+ clippings[i] = 0;
+ }
+
wi.window_type[1] = prev_type;
if (wi.window_type[0] != EIGHT_SHORT_SEQUENCE) {
+ int clipping = 0;
+
wi.num_windows = 1;
wi.grouping[0] = 1;
if (wi.window_type[0] == LONG_START_SEQUENCE)
wi.window_shape = 0;
else
wi.window_shape = 1;
+
+ for (i = 0; i < 8 && !clipping; i++)
+ clipping = clippings[i];
+ wi.clipping[0] = clipping;
} else {
int lastgrp = 0;
@@ -944,6 +968,13 @@ static FFPsyWindowInfo psy_lame_window(FFPsyContext *ctx, const float *audio,
lastgrp = i;
wi.grouping[lastgrp]++;
}
+
+ for (i = 0; i < 8; i += wi.grouping[i]) {
+ int w, clipping = 0;
+ for (w = 0; w < wi.grouping[i] && !clipping; w++)
+ clipping = clippings[i+w];
+ wi.clipping[i] = clipping;
+ }
}
/* Determine grouping, based on the location of the first attack, and save for
diff --git a/libavcodec/psymodel.h b/libavcodec/psymodel.h
index 2e3ab91..ae35809 100644
--- a/libavcodec/psymodel.h
+++ b/libavcodec/psymodel.h
@@ -66,6 +66,7 @@ typedef struct FFPsyWindowInfo {
int window_shape; ///< window shape (sine/KBD/whatever)
int num_windows; ///< number of windows in a frame
int grouping[8]; ///< window grouping (for e.g. AAC)
+ uint8_t clipping[8]; ///< set if a window group's samples are near clipping (for e.g. AAC)
int *window_sizes; ///< sequence of window sizes inside one frame (for eg. WMA)
} FFPsyWindowInfo;
diff --git a/tests/fate/aac.mak b/tests/fate/aac.mak
index 34823be..dec5a0e 100644
--- a/tests/fate/aac.mak
+++ b/tests/fate/aac.mak
@@ -91,7 +91,7 @@ fate-aac-aref-encode: CMD = enc_dec_pcm adts wav s16le $(REF) -strict -2 -c:a aa
fate-aac-aref-encode: CMP = stddev
fate-aac-aref-encode: REF = ./tests/data/asynth-44100-2.wav
fate-aac-aref-encode: CMP_SHIFT = -4096
-fate-aac-aref-encode: CMP_TARGET = 434
+fate-aac-aref-encode: CMP_TARGET = 594
fate-aac-aref-encode: SIZE_TOLERANCE = 2464
fate-aac-aref-encode: FUZZ = 5
--
1.8.4.5
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
