public envelope_lookup(vorbis_info _vi) {
ch = _vi.channels;
winlength = 128;
searchstep = 64; // not random
minenergy = _vi.codec_setup.psy_g_param.preecho_minenergy;
int n = winlength;
storage = 128;
cursor = _vi.codec_setup.blocksizes[1] / 2;
// mdct_win = _ogg_calloc(n,sizeof(*mdct_win));
mdct_win = new float[n];
mdct = new mdct_lookup();
mdct.mdct_init(n);
for (int i = 0; i < n; i++) {
mdct_win[i] = new Double(Math.sin(i / (n - 1.) * M_PI)).floatValue();
mdct_win[i] *= mdct_win[i];
}
band = new envelope_band[VE_BANDS];
for (int i = 0; i < VE_BANDS; i++) band[i] = new envelope_band();
band[0].begin = 2;
band[0].end = 4;
band[1].begin = 4;
band[1].end = 5;
band[2].begin = 6;
band[2].end = 6;
band[3].begin = 9;
band[3].end = 8;
band[4].begin = 13;
band[4].end = 8;
band[5].begin = 17;
band[5].end = 8;
band[6].begin = 22;
band[6].end = 8;
for (int j = 0; j < VE_BANDS; j++) {
n = band[j].end;
// band[j].window = _ogg_malloc(n*sizeof(*band[0].window));
band[j].window = new float[n];
for (int i = 0; i < n; i++) {
band[j].window[i] = new Double(Math.sin((i + .5) / n * M_PI)).floatValue();
band[j].total += band[j].window[i];
}
band[j].total = 1.0f / band[j].total;
}
// filter=_ogg_calloc(VE_BANDS*ch,sizeof(*filter));
// mark=_ogg_calloc(storage,sizeof(*mark));
filter = new envelope_filter_state[VE_BANDS * ch];
for (int i = 0; i < VE_BANDS * ch; i++) filter[i] = new envelope_filter_state();
mark = new int[storage];
}
public int _ve_amp(
vorbis_info_psy_global gi,
float[] data,
int offset1,
envelope_band[] bands,
envelope_filter_state[] filters,
int offset2,
int pos) {
int n = winlength;
int ret = 0;
int i, j;
float decay;
/* we want to have a 'minimum bar' for energy, else we're just
* basing blocks on quantization noise that outweighs the signal
* itself (for low power signals) */
float minV = minenergy;
// float *vec=alloca(n*sizeof(*vec));
float[] vec = new float[n];
/* stretch is used to gradually lengthen the number of windows
* considered prevoius-to-potential-trigger */
// local OOP variable rename stretch
int stretch_local = Math.max(VE_MINSTRETCH, stretch / 2);
float penalty = gi.stretch_penalty - (stretch / 2 - VE_MINSTRETCH);
if (penalty < 0.f) penalty = 0.f;
if (penalty > gi.stretch_penalty) penalty = gi.stretch_penalty;
/*_analysis_output_always("lpcm",seq2,data,n,0,0,totalshift+pos*ve->searchstep);*/
// window and transform
for (i = 0; i < n; i++) vec[i] = data[offset1 + i] * mdct_win[i];
mdct.mdct_forward(vec, vec);
// mdct.out( "", null );
/*_analysis_output_always("mdct",seq2,vec,n/2,0,1,0); */
/* near-DC spreading function; this has nothing to do with
* psychoacoustics, just sidelobe leakage and window size */
float temp = vec[0] * vec[0] + .7f * vec[1] * vec[1] + .2f * vec[2] * vec[2];
int ptr = filters[offset2].nearptr;
// the accumulation is regularly refreshed from scratch to avoid floating point creep
if (ptr == 0) {
decay = filters[offset2].nearDC_acc = filters[offset2].nearDC_partialacc + temp;
filters[offset2].nearDC_partialacc = temp;
} else {
decay = filters[offset2].nearDC_acc += temp;
filters[offset2].nearDC_partialacc += temp;
}
filters[offset2].nearDC_acc -= filters[offset2].nearDC[ptr];
filters[offset2].nearDC[ptr] = temp;
decay *= (1. / (VE_NEARDC + 1));
filters[offset2].nearptr++;
if (filters[offset2].nearptr >= VE_NEARDC) filters[offset2].nearptr = 0;
decay = todB(decay) * .5f - 15.f;
/* perform spreading and limiting, also smooth the spectrum. yes, the MDCT results
* in all real coefficients, but it still *behaves* like real/imaginary pairs */
for (i = 0; i < n / 2; i += 2) {
float val = vec[i] * vec[i] + vec[i + 1] * vec[i + 1];
val = todB(val) * .5f;
if (val < decay) val = decay;
if (val < minV) val = minV;
vec[i >> 1] = val;
decay -= 8.;
}
/*_analysis_output_always("spread",seq2++,vec,n/4,0,0,0);*/
// perform preecho/postecho triggering by band
for (j = 0; j < VE_BANDS; j++) {
float acc = 0.f;
float valmax;
float valmin;
// accumulate amplitude
for (i = 0; i < bands[j].end; i++) acc += vec[i + bands[j].begin] * bands[j].window[i];
acc *= bands[j].total;
// convert amplitude to delta
// local OOP variable rename
// protected variable name this
int p, _this = filters[offset2 + j].ampptr;
float postmax, postmin, premax = -99999.f, premin = 99999.f;
p = _this;
p--;
if (p < 0) p += VE_AMP;
postmax = Math.max(acc, filters[offset2 + j].ampbuf[p]);
postmin = Math.min(acc, filters[offset2 + j].ampbuf[p]);
for (i = 0; i < stretch_local; i++) {
p--;
if (p < 0) p += VE_AMP;
premax = Math.max(premax, filters[offset2 + j].ampbuf[p]);
premin = Math.min(premin, filters[offset2 + j].ampbuf[p]);
}
valmin = postmin - premin;
valmax = postmax - premax;
/*filters[j].markers[pos]=valmax;*/
filters[offset2 + j].ampbuf[_this] = acc;
filters[offset2 + j].ampptr++;
if (filters[offset2 + j].ampptr >= VE_AMP) filters[offset2 + j].ampptr = 0;
// look at min/max, decide trigger
if (valmax > gi.preecho_thresh[j] + penalty) {
ret |= 1;
ret |= 4;
}
if (valmin < gi.postecho_thresh[j] - penalty) ret |= 2;
}
return ret;
}