1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622
|
/*
LV2 plugin for pitch shifting, pitch correction, vocoding and harmonizing
singing voice
(c) Igor Brkic 2010
Phase vocoder code adapted from:
http://www.dspdimension.com/admin/pitch-shifting-using-the-ft/
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include <lv2plugin.hpp>
#include "vocproc.peg"
#include <fftw3.h>
#include <string.h> // because of memset
#include <stdlib.h>
#include <math.h>
using namespace LV2;
#define M_2PI 6.2831853071795
#define MAX_FRAME_LENGTH 4096
#define ROUND(a) ((float)((int)(a+0.5)))
class VocProc : public Plugin<VocProc> {
private:
float fSamplingFreq;
float sPitchFactor, sEffect, sOutputGain;
float cFormantVoco, cEffect;
float powerIn;
float sSwitch;
float pOffset[2];
float cAutoTune;
float *workspace, *gInFIFO, *gIn2FIFO, *gOutFIFO, *gOutputAccum;
float *window;
long fftFrameSize, overlap;
float freqOld;
double *fftTmpR;
fftw_complex *fftTmpC;
fftw_complex *fftOldC;
fftw_complex *fftCeps;
fftw_plan fftPlan;
void spectralEnvelope(float *env, fftw_complex *fft, uint32_t nframes);
float pitchFrequency(fftw_complex *block);
void setPitchFactor(float freq);
void phaseVocAnalysis(fftw_complex *block, float *gLastPhase, double freqPerBin, double expct, float *gAnaMagn, float *gAnaFreq);
void phaseVocSynthesis(fftw_complex *block, float *gSumPhase, float *gSynMagn, float *gSynFreq, double freqPerBin, double expct);
public:
VocProc(double rate);
~VocProc();
void run(uint32_t nframes);
};
static int _ = VocProc::register_class("http://hyperglitch.com/dev/VocProc");
// initialization
VocProc::VocProc(double rate) : Plugin<VocProc>(p_n_ports){
fSamplingFreq = (float)rate;
sPitchFactor = 1.0;
sEffect=0.0;
cEffect=0.0;
sOutputGain=1.0;
cFormantVoco=1;
pOffset[0]=0; pOffset[1]=0;
sSwitch=0.0;
cAutoTune=0.0;
powerIn=0;
fftFrameSize=2048; // pitch detection currently doesn't work for 1024
// and there is aliasing present for 1024 with formant correction when
// large pitch shifting factor is applied
overlap=4;
freqOld=0;
window=(float*)malloc(fftFrameSize*sizeof(float));
for(int k=0;k<fftFrameSize;k++)
window[k] = -.5*cos(M_2PI*(float)k/(float)fftFrameSize)+.5;
gInFIFO=(float*)calloc(fftFrameSize, sizeof(float));
#ifndef NO_VOCODER
gIn2FIFO=(float*)calloc(fftFrameSize, sizeof(float));
#endif
gOutFIFO=(float*)calloc(fftFrameSize, sizeof(float));
gOutputAccum=(float*)calloc(2*fftFrameSize, sizeof(float));
// FFTW stuff
fftTmpR=(double*)fftw_malloc(sizeof(double)*fftFrameSize);
fftTmpC=(fftw_complex*)fftw_malloc(sizeof(fftw_complex)*fftFrameSize);
fftOldC=(fftw_complex*)fftw_malloc(sizeof(fftw_complex)*fftFrameSize);
fftCeps=(fftw_complex*)fftw_malloc(sizeof(fftw_complex)*fftFrameSize);
}
VocProc::~VocProc(){
free(gInFIFO);
#ifndef NO_VOCODER
free(gIn2FIFO);
#endif
free(gOutFIFO);
free(gOutputAccum);
fftw_free(fftTmpR);
fftw_free(fftTmpC);
fftw_free(fftOldC);
fftw_free(fftCeps);
};
/*************************************************
here be procezzing
*************************************************/
void VocProc::run(uint32_t nframes) {
float *input0 = p(p_voice);
#ifndef NO_VOCODER
float *input1 = p(p_carrier);
#endif
float *output0 = p(p_output);
const float conversionTable[]={
0.500000, 0.529731, 0.561231, 0.594603, 0.629960, 0.667419, 0.707106, 0.749153, 0.793700, 0.840896, 0.890898, 0.943874,
1.000000, 1.059463, 1.122462, 1.189207, 1.259921, 1.334840, 1.414214, 1.498307, 1.587401, 1.681793, 1.781797, 1.887749, 2.000000
};
float freqScaling=*p(p_pitch_factor);
if(freqScaling<-12) freqScaling=-12;
if(freqScaling>12) freqScaling=12;
sPitchFactor = conversionTable[(int)(freqScaling+0.5)+12];
if(*p(p_effect)==0) cEffect=0;
else{
cEffect=1;
sEffect=*p(p_effect);
}
cFormantVoco=(int)(*p(p_fc_voc_switch)+0.5);
#ifdef NO_VOCODER
sSwitch=0.0;
#else
switch((int)(*p(p_fc_voc)+0.5)){
case 0:
sSwitch=0.0;
break;
case 1:
sSwitch=1.0;
break;
}
#endif
cAutoTune=ROUND(*p(p_pitch_correction));
static float gLastPhase[MAX_FRAME_LENGTH/2+1];
static float gSumPhase[MAX_FRAME_LENGTH/2+1];
static float gAnaFreq[MAX_FRAME_LENGTH];
static float gAnaMagn[MAX_FRAME_LENGTH];
static float gSynFreq[MAX_FRAME_LENGTH];
static float gSynMagn[MAX_FRAME_LENGTH];
static long gRover = false, gInit = false;
double freqPerBin, expct;
long i,k, index, inFifoLatency, stepSize, fftFrameSize2;
float *fPointer, *fPointer2, *fPointer3;
double *dPointer;
/* set up some handy variables */
fftFrameSize2 = fftFrameSize/2;
stepSize = fftFrameSize/overlap;
freqPerBin = (double)fSamplingFreq/(double)fftFrameSize;
expct = M_2PI*(double)stepSize/(double)fftFrameSize;
inFifoLatency = fftFrameSize-stepSize;
if (gRover == false) gRover = inFifoLatency;
/* initialize our static arrays */
if (gInit == false) {
memset(gLastPhase, 0, (MAX_FRAME_LENGTH/2+1)*sizeof(float));
memset(gSumPhase, 0, (MAX_FRAME_LENGTH/2+1)*sizeof(float));
memset(gAnaFreq, 0, MAX_FRAME_LENGTH*sizeof(float));
memset(gAnaMagn, 0, MAX_FRAME_LENGTH*sizeof(float));
gInit = true;
}
/* main processing loop */
for (i = 0; i < nframes; i++){
// As long as we have not yet collected enough data just read in
gInFIFO[gRover] = input0[i];
#ifndef NO_VOCODER
gIn2FIFO[gRover] = input1[i];
#endif
output0[i] = gOutFIFO[gRover-inFifoLatency];
gRover++;
// now we have enough data for processing
if (gRover >= fftFrameSize) {
gRover = inFifoLatency;
float tmpPower=0.0;
dPointer=fftTmpR;
fPointer=gInFIFO;
fPointer2=window;
for (k = 0; k < fftFrameSize;k++) {
*dPointer=*(fPointer++) * *(fPointer2++);
tmpPower+= *dPointer * *dPointer;
dPointer++;
}
powerIn=tmpPower/(float)fftFrameSize;
// do transform
fftPlan=fftw_plan_dft_r2c_1d(fftFrameSize, fftTmpR, fftTmpC, FFTW_ESTIMATE);
fftw_execute(fftPlan);
fftw_destroy_plan(fftPlan);
// pitch correction
if(cAutoTune){
float freq;
freq=pitchFrequency(fftTmpC);
setPitchFactor(freq);
}
memcpy(fftOldC, fftTmpC, fftFrameSize*sizeof(fftw_complex));
// pitch shifting with phase vocoder
phaseVocAnalysis(fftTmpC, gLastPhase, freqPerBin, expct, gAnaMagn, gAnaFreq);
memset(gSynMagn, 0, fftFrameSize*sizeof(float));
memset(gSynFreq, 0, fftFrameSize*sizeof(float));
for (k = 0; k <= fftFrameSize2; k++) {
index = k*sPitchFactor;
if (index <= fftFrameSize2) {
gSynMagn[index] += gAnaMagn[k];
gSynFreq[index] = gAnaFreq[k] * sPitchFactor;
if(cEffect)
gSynFreq[index] = gSynFreq[index]*sEffect + sEffect*200*(float)rand()/RAND_MAX-100;
}
}
phaseVocSynthesis(fftTmpC, gSumPhase, gSynMagn, gSynFreq, freqPerBin, expct);
// formant correction + vocoder
if(cFormantVoco){
float env1[fftFrameSize2], env2[fftFrameSize2];
#ifndef NO_VOCODER
if(sSwitch){
dPointer=fftTmpR; fPointer=gIn2FIFO; fPointer2=window;
for (k = 0; k < fftFrameSize;k++) {
*dPointer++=*(fPointer++) * *(fPointer2++);
}
fftPlan=fftw_plan_dft_r2c_1d(fftFrameSize, fftTmpR, fftTmpC, FFTW_ESTIMATE);
fftw_execute(fftPlan);
fftw_destroy_plan(fftPlan);
}
#endif
spectralEnvelope(env1, fftOldC, fftFrameSize2);
spectralEnvelope(env2, fftTmpC, fftFrameSize2);
// modify spectral envelope of spectrum in fftTmpC to look like spectral
// envelope of spectrum in fftOldC
float koef;
fPointer2=env1; fPointer3=env2;
for(k=0;k<fftFrameSize2;k++){
koef = *(fPointer2++) / (*(fPointer3++)+.02)*2;
fftTmpC[k][0] *= koef;
fftTmpC[k][1] *= koef;
}
}
// do inverse transform
fftPlan=fftw_plan_dft_c2r_1d(fftFrameSize, fftTmpC, fftTmpR, FFTW_ESTIMATE);
fftw_execute(fftPlan);
fftw_destroy_plan(fftPlan);
fPointer=gOutputAccum; dPointer=fftTmpR; fPointer2=window;
for(k=0; k < fftFrameSize; k++) {
*fPointer += 0.7 * *(dPointer++) / (fftFrameSize2*overlap) * sOutputGain * *(fPointer2++);
fPointer++;
}
memcpy(gOutFIFO, gOutputAccum, stepSize*sizeof(float));
// shift accumulator
memmove(gOutputAccum, gOutputAccum+stepSize, fftFrameSize*sizeof(float));
// move input FIFO
for (k = 0; k < inFifoLatency; k++) gInFIFO[k] = gInFIFO[k+stepSize];
#ifndef NO_VOCODER
for (k = 0; k < inFifoLatency; k++) gIn2FIFO[k] = gIn2FIFO[k+stepSize];
#endif
}
}
}
void VocProc::phaseVocAnalysis(fftw_complex *block, float *gLastPhase, double freqPerBin, double expct, float *gAnaMagn, float *gAnaFreq){
double real, imag, magn, phase, tmp;
long qpd, k;
for (k = 0; k <= fftFrameSize/2; k++) {
/* de-interlace FFT buffer */
real = block[k][0];
imag = block[k][1];
/* compute magnitude and phase */
magn = 2.*sqrt(real*real + imag*imag);
phase = atan2(imag,real);
/* compute phase difference */
tmp = phase - gLastPhase[k];
gLastPhase[k] = phase;
/* subtract expected phase difference */
tmp -= (double)k*expct;
/* map delta phase into +/- Pi interval */
qpd = tmp/M_PI;
if (qpd >= 0) qpd += qpd&1;
else qpd -= qpd&1;
tmp -= M_PI*(double)qpd;
/* get deviation from bin frequency from the +/- Pi interval */
tmp = overlap*tmp/(M_2PI);
/* compute the k-th partials' true frequency */
tmp = (double)k*freqPerBin + tmp*freqPerBin;
/* store magnitude and true frequency in analysis arrays */
gAnaMagn[k] = magn;
gAnaFreq[k] = tmp;
}
}
void VocProc::phaseVocSynthesis(fftw_complex *block, float *gSumPhase, float *gSynMagn, float *gSynFreq, double freqPerBin, double expct){
int k;
double magn, tmp, phase;
for (k = 0; k <= fftFrameSize/2; k++) {
/* get magnitude and true frequency from synthesis arrays */
magn = gSynMagn[k];
tmp = gSynFreq[k];
/* subtract bin mid frequency */
tmp -= (double)k*freqPerBin;
/* get bin deviation from freq deviation */
tmp /= freqPerBin;
/* take overlap into acnframes */
tmp = M_2PI*tmp/overlap;
/* add the overlap phase advance back in */
tmp += (double)k*expct;
/* accumulate delta phase to get bin phase */
gSumPhase[k] += tmp;
phase = gSumPhase[k];
/* get real and imag part and re-interleave */
block[k][0] = magn*cos(phase);
block[k][1] = magn*sin(phase);
}
}
void VocProc::spectralEnvelope(float *env, fftw_complex *fft, uint32_t nframes){
int nTaps=20;
int nTaps2=10;
float tmp[nframes+nTaps];
float h[]={ // h=(firls(20, [0 0.02 0.1 1], [1 1 0 0]));
0.0180, 0.0243, 0.0310, 0.0378, 0.0445, 0.0508, 0.0564, 0.0611,
0.0646, 0.0667, 0.0675, 0.0667, 0.0646, 0.0611, 0.0564, 0.0508,
0.0445, 0.0378, 0.0310, 0.0243, 0.0180
};
// |H(w)|
memset(tmp+nframes, 0, nTaps);
for(int k=0;k<nframes;k++)
tmp[k]=sqrt(fft[k][0]*fft[k][0]+fft[k][1]*fft[k][1]);
memset(env, 0, nframes*sizeof(float));
// magnitude spectrum filtering
int i, j;
float *p_h, *p_z, accum;
float z[2 * nTaps];
memset(z, 0, 2*nTaps*sizeof(float));
int state = 0;
for (j = 0; j < nframes+nTaps2; j++) {
z[state] = z[state + nTaps] = tmp[j];
p_h = h;
p_z = z + state;
accum = 0;
for (i = 0; i < nTaps; i++) accum += *p_h++ * *p_z++;
if (--state < 0) state += nTaps;
if(j>=nTaps2) env[j-nTaps2]=accum;
}
}
float VocProc::pitchFrequency(fftw_complex *block){
// cepstral method - needs improvement but good for now
float freq;
int i;
double cepst[fftFrameSize/2];
int ppMin, ppMax;
double max1;
float pitch=0;
//double fftTmpReal[fftFrameSize];
//fftw_complex fftTmpCplx[fftFrameSize];
for(i=0;i<fftFrameSize/2;i++){
fftCeps[i][0]=log(sqrt(pow(block[i][0], 2)+pow(block[i][1], 2))+1e-6)/(float)fftFrameSize;
fftCeps[i][1]=0;
}
fftPlan=fftw_plan_dft_c2r_1d(fftFrameSize, fftCeps, fftTmpR, FFTW_ESTIMATE);
fftw_execute(fftPlan);
fftw_destroy_plan(fftPlan);
// normalize
for(i=0;i<fftFrameSize/2;i++) cepst[i]=fabs(fftTmpR[i]/(float)fftFrameSize)+1e6;
ppMax=fftFrameSize/2-2;
ppMin=fSamplingFreq/1200; // fMax=1200
// find maximum of HTP cepstrum
max1=0;
for(i=ppMin;i<=ppMax;i++){
if(cepst[i]>max1){
max1=cepst[i];
pitch=i;
}
}
// interpolate between two samples
int idx=pitch;
int l1;
if(cepst[idx-1]>cepst[idx+1]) l1=pitch-1;
else l1=pitch;
pitch=l1+1/(cepst[l1]/cepst[l1+1]+1);
freq=fSamplingFreq/pitch;
return freq;
}
void VocProc::setPitchFactor(float freq){
// find nearest tone from chosen scale
float scale[14]; // full chromatic + edges
int scaleLen;
float mult;
// fill scale
scaleLen=1;
if(*p(p_c )==1) scale[scaleLen++]=130.813;
if(*p(p_cc)==1) scale[scaleLen++]=138.591;
if(*p(p_d )==1) scale[scaleLen++]=146.832;
if(*p(p_dd)==1) scale[scaleLen++]=155.563;
if(*p(p_e )==1) scale[scaleLen++]=164.814;
if(*p(p_f )==1) scale[scaleLen++]=174.614;
if(*p(p_ff)==1) scale[scaleLen++]=184.997;
if(*p(p_g )==1) scale[scaleLen++]=195.998;
if(*p(p_gg)==1) scale[scaleLen++]=207.652;
if(*p(p_a )==1) scale[scaleLen++]=220.000;
if(*p(p_aa)==1) scale[scaleLen++]=233.082;
if(*p(p_b )==1) scale[scaleLen++]=246.942;
if(scaleLen==1){
// no tones have been chosen - do nothing
sPitchFactor=1;
}
else{
// add edges
scale[0]=scale[scaleLen-1]/2.0;
scale[scaleLen]=scale[1]*2.0;
// calculate multiplicator
if(scale[scaleLen-1]<freq)
mult=(int)(freq/scale[scaleLen-1])+1.0;
else if(scale[1]>freq)
mult=1.0/((int)(scale[1]/freq)+1.0);
else
mult=1;
float *curr=NULL;
int i;
for(i=1;i<=scaleLen;i++){
curr=&scale[i];
if(freq<scale[i]) break;
}
// add transposition if needed (and if possible)
if((i+*p(p_transpose))<=scaleLen && (i+*p(p_transpose))>=0){
curr+=(int)(*p(p_transpose)+0.5);
}
// add hysteresis to reduce oscillation around middle point
float sign;
if((freqOld-freq)>0) sign=-1; // pitch is falling
else sign=1; // pitch is rising
float thr=(*(curr-1)+*curr)/2 + 0.3*sign*(*curr-*(curr-1));
// if current frequency is lower than threshold use lower note
if(freq<thr) curr--;
// calculate pitch factor and do pitch factor averaging (attack)
float factor=sPitchFactor;
factor*=(1+(float)((int)(*p(p_attack)*20.0)));
factor+=(*curr/freq);
factor/=((float)((int)(*p(p_attack)*20.0))+2.0);
// calculate offset (in cents)
float offset=3986*log10(factor);
if(offset<-100) offset=-100;
if(offset>100) offset=100;
if(powerIn<0.001) offset=0;
// median (this probably could be much nicer written)
float tmpSort[3];
tmpSort[0]=pOffset[0];
tmpSort[1]=pOffset[1];
tmpSort[2]=offset;
float tmp;
if(tmpSort[0]>tmpSort[1]){
tmp=tmpSort[0];
tmpSort[0]=tmpSort[1];
tmpSort[1]=tmp;
}
if(tmpSort[0]>tmpSort[2]){
tmp=tmpSort[0];
tmpSort[0]=tmpSort[2];
tmpSort[2]=tmp;
}
if(tmpSort[1]>tmpSort[2]){
tmp=tmpSort[1];
tmpSort[1]=tmpSort[2];
tmpSort[2]=tmp;
}
*p(p_offset)=tmpSort[1];
pOffset[0]=pOffset[1];
pOffset[1]=offset;
// see if correction is needed at all
if((fabs(*curr-freq)/freq)>*p(p_threshold)*0.07)
sPitchFactor=factor;
else
sPitchFactor=1;
// honor limits
if(sPitchFactor>2.0 || sPitchFactor<0.5) sPitchFactor=1.0;
}
}
|