mirror of
https://github.com/xiph/opus.git
synced 2025-06-05 15:03:39 +00:00
Fixes some issues discovered with a log sweep
- New DC rejection filter that avoids numerical issues - Higher cap on HF dynalloc - Proper filter memory for transient detector - Prevents transient detector from detecting low-freq tones as transients
This commit is contained in:
Jean-Marc Valin
parent
37df5f1df6
commit
db13c3890f
2 changed files with 54 additions and 24 deletions
45
celt/celt.c
45
celt/celt.c
|
|
@ -179,6 +179,7 @@ struct OpusCustomEncoder {
|
|||
int consec_transient;
|
||||
AnalysisInfo analysis;
|
||||
|
||||
opus_val32 transient_mem[8];
|
||||
opus_val32 preemph_memE[2];
|
||||
opus_val32 preemph_memD[2];
|
||||
|
||||
|
|
@ -293,7 +294,8 @@ static inline opus_val16 SIG2WORD16(celt_sig x)
|
|||
}
|
||||
|
||||
static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
||||
int overlap, opus_val16 *tf_estimate, int *tf_chan, AnalysisInfo *analysis)
|
||||
int overlap, opus_val16 *tf_estimate, int *tf_chan,
|
||||
AnalysisInfo *analysis, opus_val32 *transient_mem)
|
||||
{
|
||||
int i;
|
||||
VARDECL(opus_val16, tmp);
|
||||
|
|
@ -306,7 +308,7 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
VARDECL(opus_val16, bins);
|
||||
opus_val16 T1, T2, T3, T4, T5;
|
||||
opus_val16 follower;
|
||||
opus_val16 coef[2][4] = {{-2.f, 1.f, -1.f, .5f}, {0.005f, -0.995f, -1.92, .95}};
|
||||
opus_val16 coef[2][4] = {{-2.f, 1.f, -1.f, .5f}, {-1.9995f, 1.f, -1.88375f, .9025f}};
|
||||
int filterID;
|
||||
int metric=0;
|
||||
int fmetric=0, bmetric=0;
|
||||
|
|
@ -325,8 +327,8 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
{
|
||||
for (c=0;c<C;c++)
|
||||
{
|
||||
mem0=0;
|
||||
mem1=0;
|
||||
mem0 = transient_mem[filterID*4+c*2];
|
||||
mem1 = transient_mem[filterID*4+c*2+1];
|
||||
for (i=0;i<len;i++)
|
||||
tmp[i] = SHR32(in[i+c*len],SIG_SHIFT);
|
||||
|
||||
|
|
@ -336,6 +338,11 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
opus_val32 x,y;
|
||||
x = tmp[i];
|
||||
y = ADD32(mem0, x);
|
||||
if (i==len-overlap)
|
||||
{
|
||||
transient_mem[filterID*4+c*2]=mem0;
|
||||
transient_mem[filterID*4+c*2+1]=mem1;
|
||||
}
|
||||
#ifdef FIXED_POINT
|
||||
mem0 = mem1 + y - SHL32(x,1);
|
||||
mem1 = x - SHR32(y,1);
|
||||
|
|
@ -347,9 +354,6 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
#endif
|
||||
tmp[i] = EXTRACT16(SHR32(y,2));
|
||||
}
|
||||
/* First few samples are bad because we don't propagate the memory */
|
||||
for (i=0;i<12;i++)
|
||||
tmp[i] = 0;
|
||||
|
||||
maxbin=0;
|
||||
for (i=0;i<N;i++)
|
||||
|
|
@ -358,7 +362,6 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
opus_val16 max_abs=0;
|
||||
for (j=0;j<2*block;j++)
|
||||
max_abs = MAX16(max_abs, ABS16(tmp[i*block+j]));
|
||||
//printf("%f ", max_abs);
|
||||
bins[i] = max_abs;
|
||||
maxbin = MAX16(maxbin, bins[i]);
|
||||
}
|
||||
|
|
@ -412,7 +415,7 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
bmetric = 5*count1 + 4*count2 + 3*count3 + 2*count4 + count5;
|
||||
metric = fmetric+bmetric;
|
||||
|
||||
//if (metric>40)
|
||||
/*if (metric>40)*/
|
||||
if (metric>30+20*MAX16(analysis->tonality, analysis->noisiness))
|
||||
is_transient=1;
|
||||
|
||||
|
|
@ -428,7 +431,7 @@ static int transient_analysis(const opus_val32 * restrict in, int len, int C,
|
|||
#ifdef FUZZING
|
||||
is_transient = rand()&0x1;
|
||||
#endif
|
||||
//printf("%d %f %f %f %f\n", is_transient, *tf_estimate, tf_max, analysis->tonality, analysis->noisiness);
|
||||
/*printf("%d %f %d %f %f\n", is_transient, *tf_estimate, tf_max, analysis->tonality, analysis->noisiness);*/
|
||||
return is_transient;
|
||||
}
|
||||
|
||||
|
|
@ -681,7 +684,7 @@ static int tf_analysis(const CELTMode *m, int len, int C, int isTransient,
|
|||
metric[i] = best_level;
|
||||
else
|
||||
metric[i] = -best_level;
|
||||
//printf("%d ", metric[i]);
|
||||
/*printf("%d ", metric[i]);*/
|
||||
*tf_sum += (isTransient ? LM : 0) - metric[i];
|
||||
}
|
||||
/*printf("\n");*/
|
||||
|
|
@ -730,7 +733,7 @@ static int tf_analysis(const CELTMode *m, int len, int C, int isTransient,
|
|||
else
|
||||
tf_res[i] = path0[i+1];
|
||||
}
|
||||
//printf("%d %f\n", *tf_sum, tf_estimate);
|
||||
/*printf("%d %f\n", *tf_sum, tf_estimate);*/
|
||||
RESTORE_STACK;
|
||||
#ifdef FUZZING
|
||||
tf_select = rand()&0x1;
|
||||
|
|
@ -778,7 +781,7 @@ static void tf_encode(int start, int end, int isTransient, int *tf_res, int LM,
|
|||
tf_select = 0;
|
||||
for (i=start;i<end;i++)
|
||||
tf_res[i] = tf_select_table[LM][4*isTransient+2*tf_select+tf_res[i]];
|
||||
//for(i=0;i<end;i++)printf("%d ", isTransient ? LM-tf_res[i] : -tf_res[i]);printf("\n");
|
||||
/*for(i=0;i<end;i++)printf("%d ", isTransient ? LM-tf_res[i] : -tf_res[i]);printf("\n");*/
|
||||
}
|
||||
|
||||
static void tf_decode(int start, int end, int isTransient, int *tf_res, int LM, ec_dec *dec)
|
||||
|
|
@ -1207,7 +1210,7 @@ int celt_encode_with_ec(CELTEncoder * restrict st, const opus_val16 * pcm, int f
|
|||
if ((gain1 > QCONST16(.4f,15) || st->prefilter_gain > QCONST16(.4f,15)) && st->analysis.tonality > .3
|
||||
&& (pitch_index > 1.26*st->prefilter_period || pitch_index < .79*st->prefilter_period))
|
||||
pitch_change = 1;
|
||||
//printf("%d %d %f %f\n", pitch_change, pitch_index, gain1, st->analysis.tonality);
|
||||
/*printf("%d %d %f %f\n", pitch_change, pitch_index, gain1, st->analysis.tonality);*/
|
||||
if (st->loss_rate>2)
|
||||
gain1 = HALF32(gain1);
|
||||
if (st->loss_rate>4)
|
||||
|
|
@ -1306,7 +1309,7 @@ int celt_encode_with_ec(CELTEncoder * restrict st, const opus_val16 * pcm, int f
|
|||
if (st->complexity > 1)
|
||||
{
|
||||
isTransient = transient_analysis(in, N+st->overlap, CC,
|
||||
st->overlap, &tf_estimate, &tf_chan, &st->analysis);
|
||||
st->overlap, &tf_estimate, &tf_chan, &st->analysis, st->transient_mem);
|
||||
if (isTransient)
|
||||
shortBlocks = M;
|
||||
}
|
||||
|
|
@ -1435,33 +1438,33 @@ int celt_encode_with_ec(CELTEncoder * restrict st, const opus_val16 * pcm, int f
|
|||
follower[c*st->mode->nbEBands] = bandLogE2[c*st->mode->nbEBands];
|
||||
for (i=1;i<st->mode->nbEBands;i++)
|
||||
follower[c*st->mode->nbEBands+i] = MIN16(follower[c*st->mode->nbEBands+i-1]+1.5, bandLogE2[c*st->mode->nbEBands+i]);
|
||||
for (i=st->mode->nbEBands-2;i>=0;i--)
|
||||
for (i=st->end-2;i>=0;i--)
|
||||
follower[c*st->mode->nbEBands+i] = MIN16(follower[c*st->mode->nbEBands+i], MIN16(follower[c*st->mode->nbEBands+i+1]+2, bandLogE2[c*st->mode->nbEBands+i]));
|
||||
} while (++c<2);
|
||||
if (C==2)
|
||||
{
|
||||
for (i=st->start;i<st->end-1;i++)
|
||||
for (i=st->start;i<st->end;i++)
|
||||
{
|
||||
follower[st->mode->nbEBands+i] = MAX16(follower[st->mode->nbEBands+i], follower[ i]-4);
|
||||
follower[ i] = MAX16(follower[ i], follower[st->mode->nbEBands+i]-4);
|
||||
follower[i] = HALF16(MAX16(0, bandLogE[i]-follower[i]) + MAX16(0, bandLogE[st->mode->nbEBands+i]-follower[st->mode->nbEBands+i]));
|
||||
}
|
||||
} else {
|
||||
for (i=st->start;i<st->end-1;i++)
|
||||
for (i=st->start;i<st->end;i++)
|
||||
{
|
||||
follower[i] = MAX16(0, bandLogE[i]-follower[i]);
|
||||
}
|
||||
}
|
||||
for (i=st->start;i<st->end-1;i++)
|
||||
for (i=st->start;i<st->end;i++)
|
||||
{
|
||||
int width;
|
||||
int boost;
|
||||
|
||||
follower[i] = MIN16(follower[i], QCONST16(2, DB_SHIFT));
|
||||
if (i<8)
|
||||
follower[i] *= 2;
|
||||
if (i>=12)
|
||||
follower[i] *= .5;
|
||||
follower[i] = MIN16(follower[i], QCONST16(4, DB_SHIFT));
|
||||
width = C*(st->mode->eBands[i+1]-st->mode->eBands[i])<<LM;
|
||||
if (width<6)
|
||||
{
|
||||
|
|
@ -1676,7 +1679,7 @@ int celt_encode_with_ec(CELTEncoder * restrict st, const opus_val16 * pcm, int f
|
|||
/*printf ("+%d\n", adjust);*/
|
||||
}
|
||||
nbCompressedBytes = IMIN(nbCompressedBytes,nbAvailableBytes+nbFilledBytes);
|
||||
//printf("%d\n", nbCompressedBytes*50*8);
|
||||
/*printf("%d\n", nbCompressedBytes*50*8);*/
|
||||
/* This moves the raw bits to take into account the new compressed size */
|
||||
ec_enc_shrink(enc, nbCompressedBytes);
|
||||
}
|
||||
|
|
|
|||
|
|
@ -368,6 +368,29 @@ static void hp_cutoff(const opus_val16 *in, opus_int32 cutoff_Hz, opus_val16 *ou
|
|||
#endif
|
||||
}
|
||||
|
||||
static void dc_reject(const opus_val16 *in, opus_int32 cutoff_Hz, opus_val16 *out, opus_val32 *hp_mem, int len, int channels, opus_int32 Fs)
|
||||
{
|
||||
int c, i;
|
||||
float coef;
|
||||
|
||||
coef = 4.*cutoff_Hz/Fs;
|
||||
for (c=0;c<channels;c++)
|
||||
{
|
||||
for (i=0;i<len;i++)
|
||||
{
|
||||
opus_val32 x, tmp, y;
|
||||
x = in[channels*i+c];
|
||||
/* First stage */
|
||||
tmp = x-hp_mem[2*c];
|
||||
hp_mem[2*c] = hp_mem[2*c] + coef*(x - hp_mem[2*c]);
|
||||
/* Second stage */
|
||||
y = tmp - hp_mem[2*c+1];
|
||||
hp_mem[2*c+1] = hp_mem[2*c+1] + coef*(tmp - hp_mem[2*c+1]);
|
||||
out[channels*i+c] = y;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
static void stereo_fade(const opus_val16 *in, opus_val16 *out, opus_val16 g1, opus_val16 g2,
|
||||
int overlap48, int frame_size, int channels, const opus_val16 *window, opus_int32 Fs)
|
||||
{
|
||||
|
|
@ -859,9 +882,13 @@ opus_int32 opus_encode_float(OpusEncoder *st, const opus_val16 *pcm, int frame_s
|
|||
{
|
||||
hp_cutoff(pcm, cutoff_Hz, &pcm_buf[total_buffer*st->channels], st->hp_mem, frame_size, st->channels, st->Fs);
|
||||
} else {
|
||||
hp_cutoff(pcm, 3, &pcm_buf[total_buffer*st->channels], st->hp_mem, frame_size, st->channels, st->Fs);
|
||||
/*for (i=0;i<frame_size*st->channels;i++)
|
||||
pcm_buf[total_buffer*st->channels + i] = pcm[i];*/
|
||||
#ifdef FIXED_POINT
|
||||
/* FIXME: Replace by a fixed-point version of dc_reject */
|
||||
for (i=0;i<frame_size*st->channels;i++)
|
||||
pcm_buf[total_buffer*st->channels + i] = pcm[i];
|
||||
#else
|
||||
dc_reject(pcm, 3, &pcm_buf[total_buffer*st->channels], st->hp_mem, frame_size, st->channels, st->Fs);
|
||||
#endif
|
||||
}
|
||||
|
||||
#ifndef FIXED_POINT
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue