#include "fft.h"
#include "aacps.h"
#include "sbrdsp.h"
+#include "libavutil/internal.h"
#include "libavutil/libm.h"
#include <stdint.h>
static VLC vlc_sbr[10];
static const int8_t vlc_sbr_lav[10] =
{ 60, 60, 24, 24, 31, 31, 12, 12, 31, 12 };
-static const DECLARE_ALIGNED(16, float, zero64)[64];
#define SBR_INIT_VLC_STATIC(num, size) \
INIT_VLC_STATIC(&vlc_sbr[num], 9, sbr_tmp[num].table_size / sbr_tmp[num].elem_size, \
ff_ps_init();
}
+/** Places SBR in pure upsampling mode. */
+static void sbr_turnoff(SpectralBandReplication *sbr) {
+ sbr->start = 0;
+ // Init defults used in pure upsampling mode
+ sbr->kx[1] = 32; //Typo in spec, kx' inits to 32
+ sbr->m[1] = 0;
+ // Reset values for first SBR header
+ sbr->data[0].e_a[1] = sbr->data[1].e_a[1] = -1;
+ memset(&sbr->spectrum_params, -1, sizeof(SpectrumParameters));
+}
+
av_cold void ff_aac_sbr_ctx_init(AACContext *ac, SpectralBandReplication *sbr)
{
- float mdct_scale;
- sbr->kx[0] = sbr->kx[1] = 32; //Typo in spec, kx' inits to 32
- sbr->data[0].e_a[1] = sbr->data[1].e_a[1] = -1;
+ sbr->kx[0] = sbr->kx[1];
+ sbr_turnoff(sbr);
sbr->data[0].synthesis_filterbank_samples_offset = SBR_SYNTHESIS_BUF_SIZE - (1280 - 128);
sbr->data[1].synthesis_filterbank_samples_offset = SBR_SYNTHESIS_BUF_SIZE - (1280 - 128);
/* SBR requires samples to be scaled to +/-32768.0 to work correctly.
* mdct scale factors are adjusted to scale up from +/-1.0 at analysis
* and scale back down at synthesis. */
- mdct_scale = ac->avctx->sample_fmt == AV_SAMPLE_FMT_FLT ? 32768.0f : 1.0f;
- ff_mdct_init(&sbr->mdct, 7, 1, 1.0 / (64 * mdct_scale));
- ff_mdct_init(&sbr->mdct_ana, 7, 1, -2.0 * mdct_scale);
+ ff_mdct_init(&sbr->mdct, 7, 1, 1.0 / (64 * 32768.0));
+ ff_mdct_init(&sbr->mdct_ana, 7, 1, -2.0 * 32768.0);
ff_ps_ctx_init(&sbr->ps);
ff_sbrdsp_init(&sbr->dsp);
}
{
switch (bs_extension_id) {
case EXTENSION_ID_PS:
- if (!ac->m4ac.ps) {
+ if (!ac->oc[1].m4ac.ps) {
av_log(ac->avctx, AV_LOG_ERROR, "Parametric Stereo signaled to be not-present but was found in the bitstream.\n");
skip_bits_long(gb, *num_bits_left); // bs_fill_bits
*num_bits_left = 0;
} else {
#if 1
*num_bits_left -= ff_ps_read_data(ac->avctx, gb, &sbr->ps, *num_bits_left);
+ ac->avctx->profile = FF_PROFILE_AAC_HE_V2;
#else
- av_log_missing_feature(ac->avctx, "Parametric Stereo is", 0);
+ avpriv_report_missing_feature(ac->avctx, "Parametric Stereo");
skip_bits_long(gb, *num_bits_left); // bs_fill_bits
*num_bits_left = 0;
#endif
}
break;
default:
- av_log_missing_feature(ac->avctx, "Reserved SBR extensions are", 1);
+ avpriv_request_sample(ac->avctx, "Reserved SBR extensions");
skip_bits_long(gb, *num_bits_left); // bs_fill_bits
*num_bits_left = 0;
break;
if (id_aac == TYPE_SCE || id_aac == TYPE_CCE) {
if (read_sbr_single_channel_element(ac, sbr, gb)) {
- sbr->start = 0;
+ sbr_turnoff(sbr);
return get_bits_count(gb) - cnt;
}
} else if (id_aac == TYPE_CPE) {
if (read_sbr_channel_pair_element(ac, sbr, gb)) {
- sbr->start = 0;
+ sbr_turnoff(sbr);
return get_bits_count(gb) - cnt;
}
} else {
av_log(ac->avctx, AV_LOG_ERROR,
"Invalid bitstream - cannot apply SBR to element type %d\n", id_aac);
- sbr->start = 0;
+ sbr_turnoff(sbr);
return get_bits_count(gb) - cnt;
}
if (get_bits1(gb)) { // bs_extended_data
if (err < 0) {
av_log(ac->avctx, AV_LOG_ERROR,
"SBR reset failed. Switching SBR to pure upsampling mode.\n");
- sbr->start = 0;
+ sbr_turnoff(sbr);
}
}
sbr->reset = 0;
if (!sbr->sample_rate)
- sbr->sample_rate = 2 * ac->m4ac.sample_rate; //TODO use the nominal sample rate for arbitrary sample rate support
- if (!ac->m4ac.ext_sample_rate)
- ac->m4ac.ext_sample_rate = 2 * ac->m4ac.sample_rate;
+ sbr->sample_rate = 2 * ac->oc[1].m4ac.sample_rate; //TODO use the nominal sample rate for arbitrary sample rate support
+ if (!ac->oc[1].m4ac.ext_sample_rate)
+ ac->oc[1].m4ac.ext_sample_rate = 2 * ac->oc[1].m4ac.sample_rate;
if (crc) {
skip_bits(gb, 10); // bs_sbr_crc_bits; TODO - implement CRC check
//Save some state from the previous frame.
sbr->kx[0] = sbr->kx[1];
sbr->m[0] = sbr->m[1];
+ sbr->kx_and_m_pushed = 1;
num_sbr_bits++;
if (get_bits1(gb)) // bs_header_flag
* @param x pointer to the beginning of the first sample window
* @param W array of complex-valued samples split into subbands
*/
-static void sbr_qmf_analysis(DSPContext *dsp, FFTContext *mdct,
+static void sbr_qmf_analysis(AVFloatDSPContext *dsp, FFTContext *mdct,
SBRDSPContext *sbrdsp, const float *in, float *x,
- float z[320], float W[2][32][32][2])
+ float z[320], float W[2][32][32][2], int buf_idx)
{
int i;
- memcpy(W[0], W[1], sizeof(W[0]));
memcpy(x , x+1024, (320-32)*sizeof(x[0]));
memcpy(x+288, in, 1024*sizeof(x[0]));
for (i = 0; i < 32; i++) { // numTimeSlots*RATE = 16*2 as 960 sample frames
sbrdsp->sum64x5(z);
sbrdsp->qmf_pre_shuffle(z);
mdct->imdct_half(mdct, z, z+64);
- sbrdsp->qmf_post_shuffle(W[1][i], z);
+ sbrdsp->qmf_post_shuffle(W[buf_idx][i], z);
x += 32;
}
}
* Synthesis QMF Bank (14496-3 sp04 p206) and Downsampled Synthesis QMF Bank
* (14496-3 sp04 p206)
*/
-static void sbr_qmf_synthesis(DSPContext *dsp, FFTContext *mdct,
- SBRDSPContext *sbrdsp,
+static void sbr_qmf_synthesis(FFTContext *mdct,
+ SBRDSPContext *sbrdsp, AVFloatDSPContext *dsp,
float *out, float X[2][38][64],
float mdct_buf[2][64],
float *v0, int *v_off, const unsigned int div)
mdct->imdct_half(mdct, mdct_buf[1], X[1][i]);
sbrdsp->qmf_deint_bfly(v, mdct_buf[1], mdct_buf[0]);
}
- dsp->vector_fmul_add(out, v , sbr_qmf_window , zero64, 64 >> div);
+ dsp->vector_fmul (out, v , sbr_qmf_window , 64 >> div);
dsp->vector_fmul_add(out, v + ( 192 >> div), sbr_qmf_window + ( 64 >> div), out , 64 >> div);
dsp->vector_fmul_add(out, v + ( 256 >> div), sbr_qmf_window + (128 >> div), out , 64 >> div);
dsp->vector_fmul_add(out, v + ( 448 >> div), sbr_qmf_window + (192 >> div), out , 64 >> div);
(phi[1][1][0] * phi[1][1][0] + phi[1][1][1] * phi[1][1][1]) / 1.000001f;
if (!dk) {
- AV_ZERO64(alpha1[k]);
+ alpha1[k][0] = 0;
+ alpha1[k][1] = 0;
} else {
float temp_real, temp_im;
temp_real = phi[0][0][0] * phi[1][1][0] -
}
if (!phi[1][0][0]) {
- AV_ZERO64(alpha0[k]);
+ alpha0[k][0] = 0;
+ alpha0[k][1] = 0;
} else {
float temp_real, temp_im;
temp_real = phi[0][0][0] + alpha1[k][0] * phi[1][1][0] +
if (alpha1[k][0] * alpha1[k][0] + alpha1[k][1] * alpha1[k][1] >= 16.0f ||
alpha0[k][0] * alpha0[k][0] + alpha0[k][1] * alpha0[k][1] >= 16.0f) {
- AV_ZERO64(alpha0[k]);
- AV_ZERO64(alpha1[k]);
+ alpha1[k][0] = 0;
+ alpha1[k][1] = 0;
+ alpha0[k][0] = 0;
+ alpha0[k][1] = 0;
}
}
}
/// Generate the subband filtered lowband
static int sbr_lf_gen(AACContext *ac, SpectralBandReplication *sbr,
- float X_low[32][40][2], const float W[2][32][32][2])
+ float X_low[32][40][2], const float W[2][32][32][2],
+ int buf_idx)
{
int i, k;
const int t_HFGen = 8;
memset(X_low, 0, 32*sizeof(*X_low));
for (k = 0; k < sbr->kx[1]; k++) {
for (i = t_HFGen; i < i_f + t_HFGen; i++) {
- AV_COPY64(X_low[k][i], W[1][i - t_HFGen][k]);
+ X_low[k][i][0] = W[buf_idx][i - t_HFGen][k][0];
+ X_low[k][i][1] = W[buf_idx][i - t_HFGen][k][1];
}
}
+ buf_idx = 1-buf_idx;
for (k = 0; k < sbr->kx[0]; k++) {
for (i = 0; i < t_HFGen; i++) {
- AV_COPY64(X_low[k][i], W[0][i + i_f - t_HFGen][k]);
+ X_low[k][i][0] = W[buf_idx][i + i_f - t_HFGen][k][0];
+ X_low[k][i][1] = W[buf_idx][i + i_f - t_HFGen][k][1];
}
}
return 0;
/** High Frequency Adjustment (14496-3 sp04 p217) and Mapping
* (14496-3 sp04 p217)
*/
-static void sbr_mapping(AACContext *ac, SpectralBandReplication *sbr,
+static int sbr_mapping(AACContext *ac, SpectralBandReplication *sbr,
SBRData *ch_data, int e_a[2])
{
int e, i, m;
uint16_t *table = ch_data->bs_freq_res[e + 1] ? sbr->f_tablehigh : sbr->f_tablelow;
int k;
+ if (sbr->kx[1] != table[0]) {
+ av_log(ac->avctx, AV_LOG_ERROR, "kx != f_table{high,low}[0]. "
+ "Derived frequency tables were not regenerated.\n");
+ sbr_turnoff(sbr);
+ return AVERROR_BUG;
+ }
for (i = 0; i < ilim; i++)
for (m = table[i]; m < table[i + 1]; m++)
sbr->e_origmapped[e][m - sbr->kx[1]] = ch_data->env_facs[e+1][i];
}
memcpy(ch_data->s_indexmapped[0], ch_data->s_indexmapped[ch_data->bs_num_env], sizeof(ch_data->s_indexmapped[0]));
+ return 0;
}
/// Estimation of current envelope (14496-3 sp04 p218)
void ff_sbr_apply(AACContext *ac, SpectralBandReplication *sbr, int id_aac,
float* L, float* R)
{
- int downsampled = ac->m4ac.ext_sample_rate < sbr->sample_rate;
+ int downsampled = ac->oc[1].m4ac.ext_sample_rate < sbr->sample_rate;
int ch;
int nch = (id_aac == TYPE_CPE) ? 2 : 1;
+ int err;
+
+ if (!sbr->kx_and_m_pushed) {
+ sbr->kx[0] = sbr->kx[1];
+ sbr->m[0] = sbr->m[1];
+ } else {
+ sbr->kx_and_m_pushed = 0;
+ }
if (sbr->start) {
sbr_dequant(sbr, id_aac);
}
for (ch = 0; ch < nch; ch++) {
/* decode channel */
- sbr_qmf_analysis(&ac->dsp, &sbr->mdct_ana, &sbr->dsp, ch ? R : L, sbr->data[ch].analysis_filterbank_samples,
+ sbr_qmf_analysis(&ac->fdsp, &sbr->mdct_ana, &sbr->dsp, ch ? R : L, sbr->data[ch].analysis_filterbank_samples,
(float*)sbr->qmf_filter_scratch,
- sbr->data[ch].W);
- sbr_lf_gen(ac, sbr, sbr->X_low, sbr->data[ch].W);
+ sbr->data[ch].W, sbr->data[ch].Ypos);
+ sbr_lf_gen(ac, sbr, sbr->X_low,
+ (const float (*)[32][32][2]) sbr->data[ch].W,
+ sbr->data[ch].Ypos);
+ sbr->data[ch].Ypos ^= 1;
if (sbr->start) {
- sbr_hf_inverse_filter(&sbr->dsp, sbr->alpha0, sbr->alpha1, sbr->X_low, sbr->k[0]);
+ sbr_hf_inverse_filter(&sbr->dsp, sbr->alpha0, sbr->alpha1,
+ (const float (*)[40][2]) sbr->X_low, sbr->k[0]);
sbr_chirp(sbr, &sbr->data[ch]);
- sbr_hf_gen(ac, sbr, sbr->X_high, sbr->X_low, sbr->alpha0, sbr->alpha1,
+ sbr_hf_gen(ac, sbr, sbr->X_high,
+ (const float (*)[40][2]) sbr->X_low,
+ (const float (*)[2]) sbr->alpha0,
+ (const float (*)[2]) sbr->alpha1,
sbr->data[ch].bw_array, sbr->data[ch].t_env,
sbr->data[ch].bs_num_env);
// hf_adj
- sbr_mapping(ac, sbr, &sbr->data[ch], sbr->data[ch].e_a);
- sbr_env_estimate(sbr->e_curr, sbr->X_high, sbr, &sbr->data[ch]);
- sbr_gain_calc(ac, sbr, &sbr->data[ch], sbr->data[ch].e_a);
- sbr->data[ch].Ypos ^= 1;
- sbr_hf_assemble(sbr->data[ch].Y[sbr->data[ch].Ypos],
- sbr->X_high, sbr, &sbr->data[ch],
- sbr->data[ch].e_a);
+ err = sbr_mapping(ac, sbr, &sbr->data[ch], sbr->data[ch].e_a);
+ if (!err) {
+ sbr_env_estimate(sbr->e_curr, sbr->X_high, sbr, &sbr->data[ch]);
+ sbr_gain_calc(ac, sbr, &sbr->data[ch], sbr->data[ch].e_a);
+ sbr_hf_assemble(sbr->data[ch].Y[sbr->data[ch].Ypos],
+ (const float (*)[40][2]) sbr->X_high,
+ sbr, &sbr->data[ch],
+ sbr->data[ch].e_a);
+ }
}
/* synthesis */
sbr_x_gen(sbr, sbr->X[ch],
- sbr->data[ch].Y[1-sbr->data[ch].Ypos],
- sbr->data[ch].Y[ sbr->data[ch].Ypos],
- sbr->X_low, ch);
+ (const float (*)[64][2]) sbr->data[ch].Y[1-sbr->data[ch].Ypos],
+ (const float (*)[64][2]) sbr->data[ch].Y[ sbr->data[ch].Ypos],
+ (const float (*)[40][2]) sbr->X_low, ch);
}
- if (ac->m4ac.ps == 1) {
+ if (ac->oc[1].m4ac.ps == 1) {
if (sbr->ps.start) {
ff_ps_apply(ac->avctx, &sbr->ps, sbr->X[0], sbr->X[1], sbr->kx[1] + sbr->m[1]);
} else {
nch = 2;
}
- sbr_qmf_synthesis(&ac->dsp, &sbr->mdct, &sbr->dsp, L, sbr->X[0], sbr->qmf_filter_scratch,
+ sbr_qmf_synthesis(&sbr->mdct, &sbr->dsp, &ac->fdsp,
+ L, sbr->X[0], sbr->qmf_filter_scratch,
sbr->data[0].synthesis_filterbank_samples,
&sbr->data[0].synthesis_filterbank_samples_offset,
downsampled);
if (nch == 2)
- sbr_qmf_synthesis(&ac->dsp, &sbr->mdct, &sbr->dsp, R, sbr->X[1], sbr->qmf_filter_scratch,
+ sbr_qmf_synthesis(&sbr->mdct, &sbr->dsp, &ac->fdsp,
+ R, sbr->X[1], sbr->qmf_filter_scratch,
sbr->data[1].synthesis_filterbank_samples,
&sbr->data[1].synthesis_filterbank_samples_offset,
downsampled);
projects / ffmpeg / blobdiff