ffmpeg/libavcodec/aacenc_tns.c

216 lines
7.6 KiB
C
Raw Normal View History

/*
* AAC encoder TNS
* Copyright (C) 2015 Rostislav Pehlivanov
*
* 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
* AAC encoder temporal noise shaping
* @author Rostislav Pehlivanov ( atomnuker gmail com )
*/
#include "libavutil/libm.h"
#include "aacenc.h"
#include "aacenc_tns.h"
#include "aactab.h"
#include "aacenc_utils.h"
#include "aacenc_quantization.h"
/* Could be set to 3 to save an additional bit at the cost of little quality */
#define TNS_Q_BITS 4
/* Coefficient resolution in short windows */
#define TNS_Q_BITS_IS8 4
/* We really need the bits we save here elsewhere */
#define TNS_ENABLE_COEF_COMPRESSION
/* TNS will only be used if the LPC gain is within these margins */
#define TNS_GAIN_THRESHOLD_LOW 1.4f
#define TNS_GAIN_THRESHOLD_HIGH 1.16f*TNS_GAIN_THRESHOLD_LOW
static inline int compress_coeffs(int *coef, int order, int c_bits)
{
int i;
const int low_idx = c_bits ? 4 : 2;
const int shift_val = c_bits ? 8 : 4;
const int high_idx = c_bits ? 11 : 5;
#ifndef TNS_ENABLE_COEF_COMPRESSION
return 0;
#endif /* TNS_ENABLE_COEF_COMPRESSION */
for (i = 0; i < order; i++)
if (coef[i] >= low_idx && coef[i] <= high_idx)
return 0;
for (i = 0; i < order; i++)
coef[i] -= (coef[i] > high_idx) ? shift_val : 0;
return 1;
}
/**
* Encode TNS data.
* Coefficient compression is simply not lossless as it should be
* on any decoder tested and as such is not active.
*/
void ff_aac_encode_tns_info(AACEncContext *s, SingleChannelElement *sce)
{
TemporalNoiseShaping *tns = &sce->tns;
int i, w, filt, coef_compress = 0, coef_len;
const int is8 = sce->ics.window_sequence[0] == EIGHT_SHORT_SEQUENCE;
const int c_bits = is8 ? TNS_Q_BITS_IS8 == 4 : TNS_Q_BITS == 4;
if (!sce->tns.present)
return;
for (i = 0; i < sce->ics.num_windows; i++) {
put_bits(&s->pb, 2 - is8, sce->tns.n_filt[i]);
if (!tns->n_filt[i])
continue;
put_bits(&s->pb, 1, c_bits);
for (filt = 0; filt < tns->n_filt[i]; filt++) {
put_bits(&s->pb, 6 - 2 * is8, tns->length[i][filt]);
put_bits(&s->pb, 5 - 2 * is8, tns->order[i][filt]);
if (!tns->order[i][filt])
continue;
put_bits(&s->pb, 1, tns->direction[i][filt]);
coef_compress = compress_coeffs(tns->coef_idx[i][filt],
tns->order[i][filt], c_bits);
put_bits(&s->pb, 1, coef_compress);
coef_len = c_bits + 3 - coef_compress;
for (w = 0; w < tns->order[i][filt]; w++)
put_bits(&s->pb, coef_len, tns->coef_idx[i][filt][w]);
}
}
}
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
/* Apply TNS filter */
void ff_aac_apply_tns(AACEncContext *s, SingleChannelElement *sce)
{
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
TemporalNoiseShaping *tns = &sce->tns;
IndividualChannelStream *ics = &sce->ics;
int w, filt, m, i, top, order, bottom, start, end, size, inc;
const int mmm = FFMIN(ics->tns_max_bands, ics->max_sfb);
float lpc[TNS_MAX_ORDER];
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
for (w = 0; w < ics->num_windows; w++) {
bottom = ics->num_swb;
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
for (filt = 0; filt < tns->n_filt[w]; filt++) {
top = bottom;
bottom = FFMAX(0, top - tns->length[w][filt]);
order = tns->order[w][filt];
if (order == 0)
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
continue;
// tns_decode_coef
compute_lpc_coefs(tns->coef[w][filt], order, lpc, 0, 0, 0);
start = ics->swb_offset[FFMIN(bottom, mmm)];
end = ics->swb_offset[FFMIN( top, mmm)];
aacenc_tns: rework the way coefficients are calculated This commit abandons the way the specifications state to quantize the coefficients, makes use of the new LPC float functions and is much better. The original way of converting non-normalized float samples to int32_t which out LPC system expects was wrong and it was wrong to assume the coefficients that are generated are also valid. It was essentially a full garbage-in, garbage-out system and it definitely shows when looking at spectrals and listening. The high frequencies were very overattenuated. The new LPC function performs the analysis directly. The specifications state to quantize the coefficients into four bit index values using an asin() function which of course had to have ugly ternary operators because the function turns negative if the coefficients are negative which when encoding causes invalid bitstream to get generated. This deviates from this by using the direct TNS tables, which are fairly small since you only have 4 bits at most for index values. The LPC values are directly quantized against the tables and are then used to perform filtering after the requantization, which simply fetches the array values. The end result is that TNS works much better now and doesn't attenuate anything but the actual signal, e.g. TNS removes quantization errors and does it's job correctly now. It might be enabled by default soon since it doesn't hurt and helps reduce nastyness at low bitrates. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2015-08-29 07:47:31 +02:00
if ((size = end - start) <= 0)
continue;
if (tns->direction[w][filt]) {
inc = -1;
start = end - 1;
} else {
inc = 1;
}
start += w * 128;
/* AR filter */
for (m = 0; m < size; m++, start += inc) {
for (i = 1; i <= FFMIN(m, order); i++) {
sce->coeffs[start] += lpc[i-1]*sce->pcoeffs[start - i*inc];
}
}
}
}
}
/*
* c_bits - 1 if 4 bit coefficients, 0 if 3 bit coefficients
*/
static inline void quantize_coefs(double *coef, int *idx, float *lpc, int order,
int c_bits)
{
int i;
const float *quant_arr = tns_tmp2_map[c_bits];
for (i = 0; i < order; i++) {
idx[i] = quant_array_idx(coef[i], quant_arr, c_bits ? 16 : 8);
lpc[i] = quant_arr[idx[i]];
}
}
/*
* 3 bits per coefficient with 8 short windows
*/
void ff_aac_search_for_tns(AACEncContext *s, SingleChannelElement *sce)
{
TemporalNoiseShaping *tns = &sce->tns;
int w, g, count = 0;
double gain, coefs[MAX_LPC_ORDER];
const int mmm = FFMIN(sce->ics.tns_max_bands, sce->ics.max_sfb);
const int is8 = sce->ics.window_sequence[0] == EIGHT_SHORT_SEQUENCE;
const int c_bits = is8 ? TNS_Q_BITS_IS8 == 4 : TNS_Q_BITS == 4;
const int sfb_start = av_clip(tns_min_sfb[is8][s->samplerate_index], 0, mmm);
const int sfb_end = av_clip(sce->ics.num_swb, 0, mmm);
const int order = is8 ? 7 : s->profile == FF_PROFILE_AAC_LOW ? 12 : TNS_MAX_ORDER;
const int slant = sce->ics.window_sequence[0] == LONG_STOP_SEQUENCE ? 1 :
sce->ics.window_sequence[0] == LONG_START_SEQUENCE ? 0 : 2;
const int sfb_len = sfb_end - sfb_start;
const int coef_len = sce->ics.swb_offset[sfb_end] - sce->ics.swb_offset[sfb_start];
if (coef_len <= 0 || sfb_len <= 0) {
sce->tns.present = 0;
return;
}
for (w = 0; w < sce->ics.num_windows; w++) {
float en[2] = {0.0f, 0.0f};
int oc_start = 0, os_start = 0;
int coef_start = sce->ics.swb_offset[sfb_start];
for (g = sfb_start; g < sce->ics.num_swb && g <= sfb_end; g++) {
FFPsyBand *band = &s->psy.ch[s->cur_channel].psy_bands[w*16+g];
if (g > sfb_start + (sfb_len/2))
en[1] += band->energy;
else
en[0] += band->energy;
}
/* LPC */
gain = ff_lpc_calc_ref_coefs_f(&s->lpc, &sce->coeffs[w*128 + coef_start],
coef_len, order, coefs);
if (!order || !isfinite(gain) || gain < TNS_GAIN_THRESHOLD_LOW || gain > TNS_GAIN_THRESHOLD_HIGH)
continue;
tns->n_filt[w] = is8 ? 1 : order != TNS_MAX_ORDER ? 2 : 3;
for (g = 0; g < tns->n_filt[w]; g++) {
tns->direction[w][g] = slant != 2 ? slant : en[g] < en[!g];
tns->order[w][g] = g < tns->n_filt[w] ? order/tns->n_filt[w] : order - oc_start;
tns->length[w][g] = g < tns->n_filt[w] ? sfb_len/tns->n_filt[w] : sfb_len - os_start;
quantize_coefs(&coefs[oc_start], tns->coef_idx[w][g], tns->coef[w][g],
tns->order[w][g], c_bits);
oc_start += tns->order[w][g];
os_start += tns->length[w][g];
}
count++;
}
sce->tns.present = !!count;
}