ref: 2788a3510e621d1191849f5a03fd090de8bcba25
dir: /LEAF/Src/leaf-utilities.c/
/* ============================================================================== LEAFUtilities.c Created: 20 Jan 2017 12:02:17pm Author: Michael R Mulshine ============================================================================== */ #if _WIN32 || _WIN64 #include "..\Inc\leaf-utilities.h" #include "..\Inc\leaf-wavetables.h" #include "..\leaf.h" #include "..\Inc\d_fft_mayer.h" #else #include "../Inc/leaf-utilities.h" #include "../Inc/leaf-wavetables.h" #include "../leaf.h" #include "../Externals/d_fft_mayer.h" #endif #define LOGTEN 2.302585092994 float mtof(float f) { if (f <= -1500.0f) return(0); else if (f > 1499.0f) return(mtof(1499.0f)); else return (8.17579891564f * exp(0.0577622650f * f)); } float ftom(float f) { return (f > 0 ? 17.3123405046f * log(.12231220585f * f) : -1500.0f); } float powtodb(float f) { if (f <= 0) return (0); else { float val = 100 + 10.f/LOGTEN * log(f); return (val < 0 ? 0 : val); } } float rmstodb(float f) { if (f <= 0) return (0); else { float val = 100 + 20.f/LOGTEN * log(f); return (val < 0 ? 0 : val); } } float dbtopow(float f) { if (f <= 0) return(0); else { if (f > 870.0f) f = 870.0f; return (exp((LOGTEN * 0.1f) * (f-100.0f))); } } float dbtorms(float f) { if (f <= 0) return(0); else { if (f > 485.0f) f = 485.0f; } return (exp((LOGTEN * 0.05f) * (f-100.0f))); } /* ---------------- env~ - simple envelope follower. ----------------- */ void tEnv_init(tEnv* const x, int ws, int hs, int bs) { int period = hs, npoints = ws; int i; if (npoints < 1) npoints = 1024; if (period < 1) period = npoints/2; if (period < npoints / MAXOVERLAP + 1) period = npoints / MAXOVERLAP + 1; x->x_npoints = npoints; x->x_phase = 0; x->x_period = period; x->windowSize = npoints; x->hopSize = period; x->blockSize = bs; for (i = 0; i < MAXOVERLAP; i++) x->x_sumbuf[i] = 0; for (i = 0; i < npoints; i++) x->buf[i] = (1.0f - cos((2 * PI * i) / npoints))/npoints; for (; i < npoints+INITVSTAKEN; i++) x->buf[i] = 0; x->x_f = 0; x->x_allocforvs = INITVSTAKEN; // ~ ~ ~ dsp ~ ~ ~ if (x->x_period % x->blockSize) { x->x_realperiod = x->x_period + x->blockSize - (x->x_period % x->blockSize); } else { x->x_realperiod = x->x_period; } // ~ ~ ~ ~ ~ ~ ~ ~ } void tEnv_free (tEnv* const x) { leaf_free(x); } float tEnv_tick (tEnv* const x) { return powtodb(x->x_result); } void tEnv_processBlock(tEnv* const x, float* in) { int n = x->blockSize; int count; t_sample *sump; in += n; for (count = x->x_phase, sump = x->x_sumbuf; count < x->x_npoints; count += x->x_realperiod, sump++) { t_sample *hp = x->buf + count; t_sample *fp = in; t_sample sum = *sump; int i; for (i = 0; i < n; i++) { fp--; sum += *hp++ * (*fp * *fp); } *sump = sum; } sump[0] = 0; x->x_phase -= n; if (x->x_phase < 0) { x->x_result = x->x_sumbuf[0]; for (count = x->x_realperiod, sump = x->x_sumbuf; count < x->x_npoints; count += x->x_realperiod, sump++) sump[0] = sump[1]; sump[0] = 0; x->x_phase = x->x_realperiod - n; } } // ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Compressor ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ // /* tCompressor* tCompressorInit(int tauAttack, int tauRelease) { tCompressor* c = &leaf.tCompressorRegistry[leaf.registryIndex[T_COMPRESSOR]++]; c->tauAttack = tauAttack; c->tauRelease = tauRelease; c->x_G[0] = 0.0f, c->x_G[1] = 0.0f, c->y_G[0] = 0.0f, c->y_G[1] = 0.0f, c->x_T[0] = 0.0f, c->x_T[1] = 0.0f, c->y_T[0] = 0.0f, c->y_T[1] = 0.0f; c->T = 0.0f; // Threshold c->R = 1.0f; // compression Ratio c->M = 0.0f; // decibel Make-up gain c->W = 0.0f; // decibel Width of knee transition return c; } */ void tCompressor_init(tCompressor* const c) { c->tauAttack = 100; c->tauRelease = 100; c->isActive = OFALSE; c->T = 0.0f; // Threshold c->R = 0.5f; // compression Ratio c->M = 3.0f; // decibel Width of knee transition c->W = 1.0f; // decibel Make-up gain } int ccount = 0; float tCompressor_tick(tCompressor* const c, float in) { float slope, overshoot; float alphaAtt, alphaRel; float in_db = 20.0f * log10f( fmaxf( fabsf( in), 0.000001f)), out_db = 0.0f; c->y_T[1] = c->y_T[0]; slope = c->R - 1.0f; // feed-forward topology; was 1/C->R - 1 overshoot = in_db - c->T; if (overshoot <= -(c->W * 0.5f)) { out_db = in_db; c->isActive = OFALSE; } else if ((overshoot > -(c->W * 0.5f)) && (overshoot < (c->W * 0.5f))) { out_db = in_db + slope * (powf((overshoot + c->W*0.5f),2) / (2.0f * c->W)); // .^ 2 ??? c->isActive = OTRUE; } else if (overshoot >= (c->W * 0.5f)) { out_db = in_db + slope * overshoot; c->isActive = OTRUE; } c->x_T[0] = out_db - in_db; alphaAtt = expf(-1.0f/(0.001f * c->tauAttack * leaf.sampleRate)); alphaRel = expf(-1.0f/(0.001f * c->tauRelease * leaf.sampleRate)); if (c->x_T[0] > c->y_T[1]) c->y_T[0] = alphaAtt * c->y_T[1] + (1-alphaAtt) * c->x_T[0]; else c->y_T[0] = alphaRel * c->y_T[1] + (1-alphaRel) * c->x_T[0]; float attenuation = powf(10.0f, ((c->M - c->y_T[0])/20.0f)); /* if (++ccount > 5000) { ccount = 0; DBG(".5width: " + String(c->W * 0.5f)); DBG("slope: " + String(slope) + " overshoot: " + String(overshoot)); DBG("attenuation: " + String(attenuation)); } */ return attenuation * in; } // ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Envelope ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ // void tEnvelope_init(tEnvelope* const env, float attack, float decay, oBool loop) { env->exp_buff = exp_decay; env->inc_buff = attack_decay_inc; env->buff_size = sizeof(exp_decay); env->loop = loop; if (attack > 8192.0f) attack = 8192.0f; if (attack < 0.0f) attack = 0.0f; if (decay > 8192.0f) decay = 8192.0f; if (decay < 0.0f) decay = 0.0f; int16_t attackIndex = ((int16_t)(attack * 8.0f))-1; int16_t decayIndex = ((int16_t)(decay * 8.0f))-1; int16_t rampIndex = ((int16_t)(2.0f * 8.0f))-1; if (attackIndex < 0) attackIndex = 0; if (decayIndex < 0) decayIndex = 0; if (rampIndex < 0) rampIndex = 0; env->inRamp = OFALSE; env->inAttack = OFALSE; env->inDecay = OFALSE; env->attackInc = env->inc_buff[attackIndex]; env->decayInc = env->inc_buff[decayIndex]; env->rampInc = env->inc_buff[rampIndex]; } void tEnvelope_free(tEnvelope* const env) { leaf_free(env); } int tEnvelope_setAttack(tEnvelope* const env, float attack) { int32_t attackIndex; if (attack < 0.0f) { attackIndex = 0.0f; } else if (attack < 8192.0f) { attackIndex = ((int32_t)(attack * 8.0f))-1; } else { attackIndex = ((int32_t)(8192.0f * 8.0f))-1; } env->attackInc = env->inc_buff[attackIndex]; return 0; } int tEnvelope_setDecay(tEnvelope* const env, float decay) { int32_t decayIndex; if (decay <= 1.0f) { decayIndex = 1; } else if (decay < 8192.0f) { decayIndex = ((int32_t)(decay * 8.0f)) - 1; } else { decayIndex = ((int32_t)(8192.0f * 8.0f)) - 1; } env->decayInc = env->inc_buff[decayIndex]; return 0; } int tEnvelope_loop(tEnvelope* const env, oBool loop) { env->loop = loop; return 0; } int tEnvelope_on(tEnvelope* const env, float velocity) { if (env->inAttack || env->inDecay) // In case envelope retriggered while it is still happening. { env->rampPhase = 0; env->inRamp = OTRUE; env->rampPeak = env->next; } else // Normal start. { env->inAttack = OTRUE; } env->attackPhase = 0; env->decayPhase = 0; env->inDecay = OFALSE; env->gain = velocity; return 0; } float tEnvelope_tick(tEnvelope* const env) { if (env->inRamp) { if (env->rampPhase > UINT16_MAX) { env->inRamp = OFALSE; env->inAttack = OTRUE; env->next = 0.0f; } else { env->next = env->rampPeak * env->exp_buff[(uint32_t)env->rampPhase]; } env->rampPhase += env->rampInc; } if (env->inAttack) { // If attack done, time to turn around. if (env->attackPhase > UINT16_MAX) { env->inDecay = OTRUE; env->inAttack = OFALSE; env->next = env->gain * 1.0f; } else { // do interpolation ! env->next = env->gain * env->exp_buff[UINT16_MAX - (uint32_t)env->attackPhase]; // inverted and backwards to get proper rising exponential shape/perception } // Increment envelope attack. env->attackPhase += env->attackInc; } if (env->inDecay) { // If decay done, finish. if (env->decayPhase >= UINT16_MAX) { env->inDecay = OFALSE; if (env->loop) { env->attackPhase = 0; env->decayPhase = 0; env->inAttack = OTRUE; } else { env->next = 0.0f; } } else { env->next = env->gain * (env->exp_buff[(uint32_t)env->decayPhase]); // do interpolation ! } // Increment envelope decay; env->decayPhase += env->decayInc; } return env->next; } /* ADSR */ void tADSR_init(tADSR* const adsr, float attack, float decay, float sustain, float release) { adsr->exp_buff = exp_decay; adsr->inc_buff = attack_decay_inc; adsr->buff_size = sizeof(exp_decay); if (attack > 8192.0f) attack = 8192.0f; if (attack < 0.0f) attack = 0.0f; if (decay > 8192.0f) decay = 8192.0f; if (decay < 0.0f) decay = 0.0f; if (sustain > 1.0f) sustain = 1.0f; if (sustain < 0.0f) sustain = 0.0f; if (release > 8192.0f) release = 8192.0f; if (release < 0.0f) release = 0.0f; int16_t attackIndex = ((int16_t)(attack * 8.0f))-1; int16_t decayIndex = ((int16_t)(decay * 8.0f))-1; int16_t releaseIndex = ((int16_t)(release * 8.0f))-1; int16_t rampIndex = ((int16_t)(2.0f * 8.0f))-1; if (attackIndex < 0) attackIndex = 0; if (decayIndex < 0) decayIndex = 0; if (releaseIndex < 0) releaseIndex = 0; if (rampIndex < 0) rampIndex = 0; adsr->inRamp = OFALSE; adsr->inAttack = OFALSE; adsr->inDecay = OFALSE; adsr->inSustain = OFALSE; adsr->inRelease = OFALSE; adsr->sustain = sustain; adsr->attackInc = adsr->inc_buff[attackIndex]; adsr->decayInc = adsr->inc_buff[decayIndex]; adsr->releaseInc = adsr->inc_buff[releaseIndex]; adsr->rampInc = adsr->inc_buff[rampIndex]; } int tADSR_setAttack(tADSR* const adsr, float attack) { int32_t attackIndex; if (attack < 0.0f) { attackIndex = 0.0f; } else if (attack < 8192.0f) { attackIndex = ((int32_t)(attack * 8.0f))-1; } else { attackIndex = ((int32_t)(8192.0f * 8.0f))-1; } adsr->attackInc = adsr->inc_buff[attackIndex]; return 0; } int tADSR_detDecay(tADSR* const adsr, float decay) { int32_t decayIndex; if (decay < 0.0f) { decayIndex = 0.0f; } else if (decay < 8192.0f) { decayIndex = ((int32_t)(decay * 8.0f)) - 1; } else { decayIndex = ((int32_t)(8192.0f * 8.0f)) - 1; } adsr->decayInc = adsr->inc_buff[decayIndex]; return 0; } int tADSR_setSustain(tADSR *const adsr, float sustain) { if (sustain > 1.0f) adsr->sustain = 1.0f; else if (sustain < 0.0f) adsr->sustain = 0.0f; else adsr->sustain = sustain; return 0; } int tADSR_setRelease(tADSR* const adsr, float release) { int32_t releaseIndex; if (release < 0.0f) { releaseIndex = 0.0f; } else if (release < 8192.0f) { releaseIndex = ((int32_t)(release * 8.0f)) - 1; } else { releaseIndex = ((int32_t)(8192.0f * 8.0f)) - 1; } adsr->releaseInc = adsr->inc_buff[releaseIndex]; return 0; } int tADSR_on(tADSR* const adsr, float velocity) { if ((adsr->inAttack || adsr->inDecay) || (adsr->inSustain || adsr->inRelease)) // In case ADSR retriggered while it is still happening. { adsr->rampPhase = 0; adsr->inRamp = OTRUE; adsr->rampPeak = adsr->next; } else // Normal start. { adsr->inAttack = OTRUE; } adsr->attackPhase = 0; adsr->decayPhase = 0; adsr->releasePhase = 0; adsr->inDecay = OFALSE; adsr->inSustain = OFALSE; adsr->inRelease = OFALSE; adsr->gain = velocity; return 0; } int tADSR_off(tADSR* const adsr) { if (adsr->inRelease) return 0; adsr->inAttack = OFALSE; adsr->inDecay = OFALSE; adsr->inSustain = OFALSE; adsr->inRelease = OTRUE; adsr->releasePeak = adsr->next; return 0; } float tADSR_tick(tADSR* const adsr) { if (adsr->inRamp) { if (adsr->rampPhase > UINT16_MAX) { adsr->inRamp = OFALSE; adsr->inAttack = OTRUE; adsr->next = 0.0f; } else { adsr->next = adsr->rampPeak * adsr->exp_buff[(uint32_t)adsr->rampPhase]; } adsr->rampPhase += adsr->rampInc; } if (adsr->inAttack) { // If attack done, time to turn around. if (adsr->attackPhase > UINT16_MAX) { adsr->inDecay = OTRUE; adsr->inAttack = OFALSE; adsr->next = adsr->gain * 1.0f; } else { // do interpolation ! adsr->next = adsr->gain * adsr->exp_buff[UINT16_MAX - (uint32_t)adsr->attackPhase]; // inverted and backwards to get proper rising exponential shape/perception } // Increment ADSR attack. adsr->attackPhase += adsr->attackInc; } if (adsr->inDecay) { // If decay done, sustain. if (adsr->decayPhase >= UINT16_MAX) { adsr->inDecay = OFALSE; adsr->inSustain = OTRUE; adsr->next = adsr->gain * adsr->sustain; } else { adsr->next = adsr->gain * (adsr->sustain + ((adsr->exp_buff[(uint32_t)adsr->decayPhase]) * (1 - adsr->sustain))); // do interpolation ! } // Increment ADSR decay. adsr->decayPhase += adsr->decayInc; } if (adsr->inRelease) { // If release done, finish. if (adsr->releasePhase >= UINT16_MAX) { adsr->inRelease = OFALSE; adsr->next = 0.0f; } else { adsr->next = adsr->releasePeak * (adsr->exp_buff[(uint32_t)adsr->releasePhase]); // do interpolation ! } // Increment envelope release; adsr->releasePhase += adsr->releaseInc; } return adsr->next; } /* Envelope Follower */ void tEnvelopeFollower_init(tEnvelopeFollower* const e, float attackThreshold, float decayCoeff) { e->y = 0.0f; e->a_thresh = attackThreshold; e->d_coeff = decayCoeff; } float tEnvelopeFollower_tick(tEnvelopeFollower* const ef, float x) { if (x < 0.0f ) x = -x; /* Absolute value. */ if ((x >= ef->y) && (x > ef->a_thresh)) ef->y = x; /* If we hit a peak, ride the peak to the top. */ else ef->y = ef->y * ef->d_coeff; /* Else, exponential decay of output. */ //ef->y = envelope_pow[(uint16_t)(ef->y * (float)UINT16_MAX)] * ef->d_coeff; //not quite the right behavior - too much loss of precision? //ef->y = powf(ef->y, 1.000009f) * ef->d_coeff; // too expensive if( ef->y < VSF) ef->y = 0.0f; return ef->y; } int tEnvelopeFollower_decayCoeff(tEnvelopeFollower* const ef, float decayCoeff) { return ef->d_coeff = decayCoeff; } int tEnvelopeFollower_attackThresh(tEnvelopeFollower* const ef, float attackThresh) { return ef->a_thresh = attackThresh; } /* Ramp */ void tRamp_init(tRamp* const ramp, float time, int samples_per_tick) { ramp->inv_sr_ms = 1.0f/(leaf.sampleRate*0.001f); ramp->minimum_time = ramp->inv_sr_ms * samples_per_tick; ramp->curr = 0.0f; ramp->dest = 0.0f; if (time < ramp->minimum_time) { ramp->time = ramp->minimum_time; } else { ramp->time = time; } ramp->samples_per_tick = samples_per_tick; ramp->inc = ((ramp->dest - ramp->curr) / ramp->time * ramp->inv_sr_ms) * (float)ramp->samples_per_tick; } int tRamp_setTime(tRamp* const r, float time) { if (time < r->minimum_time) { r->time = r->minimum_time; } else { r->time = time; } r->inc = ((r->dest-r->curr)/r->time * r->inv_sr_ms) * ((float)r->samples_per_tick); return 0; } int tRamp_setDest(tRamp* const r, float dest) { r->dest = dest; r->inc = ((r->dest-r->curr)/r->time * r->inv_sr_ms) * ((float)r->samples_per_tick); return 0; } int tRamp_setVal(tRamp* const r, float val) { r->curr = val; r->inc = ((r->dest-r->curr)/r->time * r->inv_sr_ms) * ((float)r->samples_per_tick); return 0; } float tRamp_tick(tRamp* const r) { r->curr += r->inc; if (((r->curr >= r->dest) && (r->inc > 0.0f)) || ((r->curr <= r->dest) && (r->inc < 0.0f))) r->inc = 0.0f; return r->curr; } float tRamp_sample(tRamp* const r) { return r->curr; } void tRampSampleRateChanged(tRamp* const r) { r->inv_sr_ms = 1.0f / (leaf.sampleRate * 0.001f); r->inc = ((r->dest-r->curr)/r->time * r->inv_sr_ms)*((float)r->samples_per_tick); } // If stack contains note, returns index. Else returns -1; int tStack_contains(tStack* const ns, uint16_t noteVal) { for (int i = 0; i < ns->size; i++) { if (ns->data[i] == noteVal) return i; } return -1; } void tStack_add(tStack* const ns, uint16_t noteVal) { uint8_t j; int whereToInsert = 0; if (ns->ordered) { for (j = 0; j < ns->size; j++) { if (noteVal > ns->data[j]) { if ((noteVal < ns->data[j+1]) || (ns->data[j+1] == -1)) { whereToInsert = j+1; break; } } } } //first move notes that are already in the stack one position to the right for (j = ns->size; j > whereToInsert; j--) { ns->data[j] = ns->data[(j - 1)]; } //then, insert the new note into the front of the stack ns->data[whereToInsert] = noteVal; ns->size++; } int tStack_addIfNotAlreadyThere(tStack* const ns, uint16_t noteVal) { uint8_t j; int added = 0; if (tStack_contains(ns, noteVal) == -1) { int whereToInsert = 0; if (ns->ordered) { for (j = 0; j < ns->size; j++) { if (noteVal > ns->data[j]) { if ((noteVal < ns->data[j+1]) || (ns->data[j+1] == -1)) { whereToInsert = j+1; break; } } } } //first move notes that are already in the stack one position to the right for (j = ns->size; j > whereToInsert; j--) { ns->data[j] = ns->data[(j - 1)]; } //then, insert the new note into the front of the stack ns->data[whereToInsert] = noteVal; ns->size++; added = 1; } return added; } // Remove noteVal. return 1 if removed, 0 if not int tStack_remove(tStack* const ns, uint16_t noteVal) { uint8_t k; int foundIndex = tStack_contains(ns, noteVal); int removed = 0; if (foundIndex >= 0) { for (k = 0; k < (ns->size - foundIndex); k++) { if ((k+foundIndex) >= (ns->capacity - 1)) { ns->data[k + foundIndex] = -1; } else { ns->data[k + foundIndex] = ns->data[k + foundIndex + 1]; if ((k + foundIndex) == (ns->size - 1)) { ns->data[k + foundIndex + 1] = -1; } } } // in case it got put on the stack multiple times foundIndex--; ns->size--; removed = 1; } return removed; } // Doesn't change size of data types void tStack_setCapacity(tStack* const ns, uint16_t cap) { if (cap <= 0) ns->capacity = 1; else if (cap <= STACK_SIZE) ns->capacity = cap; else ns->capacity = STACK_SIZE; for (int i = cap; i < STACK_SIZE; i++) { if ((int)ns->data[i] != -1) { ns->data[i] = -1; ns->size -= 1; } } if (ns->pos >= cap) { ns->pos = 0; } } int tStack_getSize(tStack* const ns) { return ns->size; } void tStack_clear(tStack* const ns) { for (int i = 0; i < STACK_SIZE; i++) { ns->data[i] = -1; } ns->pos = 0; ns->size = 0; } // Next item in order of addition to stack. Return 0-31 if there is a next item to move to. Returns -1 otherwise. int tStack_next(tStack* const ns) { int step = 0; if (ns->size != 0) // if there is at least one note in the stack { if (ns->pos > 0) // if you're not at the most recent note (first one), then go backward in the array (moving from earliest to latest) { ns->pos--; } else { ns->pos = (ns->size - 1); // if you are the most recent note, go back to the earliest note in the array } step = ns->data[ns->pos]; return step; } else { return -1; } } int tStack_get(tStack* const ns, int which) { return ns->data[which]; } int tStack_first(tStack* const ns) { return ns->data[0]; } void tStack_init(tStack* const ns) { ns->ordered = OFALSE; ns->size = 0; ns->pos = 0; ns->capacity = STACK_SIZE; for (int i = 0; i < STACK_SIZE; i++) ns->data[i] = -1; } /******************************************************************************/ /***************** static function declarations *******************************/ /******************************************************************************/ static void solad_init(tSOLAD *w); static inline float read_sample(tSOLAD *w, float floatindex); static void pitchdown(tSOLAD *w, float *out); static void pitchup(tSOLAD *w, float *out); /******************************************************************************/ /***************** public access functions ************************************/ /******************************************************************************/ // init void tSOLAD_init(tSOLAD* const w) { w->pitchfactor = 1.; solad_init(w); } void tSOLAD_free(tSOLAD* const w) { leaf_free(w); } // send one block of input samples, receive one block of output samples void tSOLAD_ioSamples(tSOLAD* const w, float* in, float* out, int blocksize) { int i = w->timeindex; int n = w->blocksize = blocksize; if(!i) w->delaybuf[LOOPSIZE] = in[0]; // copy one sample for interpolation while(n--) w->delaybuf[i++] = *in++; // copy one input block to delay buffer if(w->pitchfactor > 1) pitchup(w, out); else pitchdown(w, out); w->timeindex += blocksize; w->timeindex &= LOOPMASK; } // set periodicity analysis data void tSOLAD_setPeriod(tSOLAD* const w, float period) { if(period > MAXPERIOD) period = MAXPERIOD; if(period > MINPERIOD) w->period = period; // ignore period when too small } // set pitch factor between 0.25 and 4 void tSOLAD_setPitchFactor(tSOLAD* const w, float pitchfactor) { if(pitchfactor < 0.25) pitchfactor = 0.25; else if(pitchfactor > 4.) pitchfactor = 4.; w->pitchfactor = pitchfactor; } // force readpointer lag void tSOLAD_setReadLag(tSOLAD* const w, float readlag) { if(readlag < 0) readlag = 0; if(readlag < w->readlag) // do not jump backward, only forward { w->jump = w->readlag - readlag; w->readlag = readlag; w->xfadelength = readlag; w->xfadevalue = 1; } } // reset state variables void tSOLAD_resetState(tSOLAD* const w) { int n = LOOPSIZE + 1; float *buf = w->delaybuf; while(n--) *buf++ = 0; solad_init(w); } /******************************************************************************/ /******************** private procedures **************************************/ /******************************************************************************/ /* Function pitchdown() is called to read samples from the delay buffer when pitch factor is between 0.25 and 1. The read pointer lags behind because of the slowed down speed, and it must jump forward towards the write pointer soon as there is sufficient space to jump. That is, if there is at least one period of the input signal between read pointer and write pointer. When short periods follow up on long periods, the read pointer may have space to jump over more than one period lenghts. Jump length must be [periodlength ^ 2] in any case. A linear crossfade function joins the jump-from point with the jump-to point. The crossfade must be completed before another read pointer jump is allowed. Length of the crossfade function is stored as a number of samples in terms of the input sample rate. This length is dynamically translated to a crossfade length expressed in output reading rate, according to pitch factor which can change before the crossfade is completed. Crossfade length does not cover an invariable length in periods for all pitch transposition factors. For pitch factors from 0.5 till 1, crossfade length is stretched in the output just as much as the signal itself, as crossfade speed is set to equal pitch factor. For pitch factors below 0.5, the read pointer wants to jump forward before one period is read, therefore the crossfade length as expressed in output periods must be shorter. Crossfade speed is set to [1 - pitchfactor] for those cases. Pitch factor 0.5 is the natural switch point between crossfade speeds [pitchfactor] and [1 - pitchfactor] because 0.5 == 1 - 0.5. The crossfade speed modification for pitch factors below 0.5 also means that much of the original signal content will be skipped. */ static void pitchdown(tSOLAD* const w, float *out) { int n = w->blocksize; float refindex = (float)(w->timeindex + LOOPSIZE); // no negative values! float pitchfactor = w->pitchfactor; float period = w->period; float readlag = w->readlag; float readlagstep = 1 - pitchfactor; float jump = w->jump; float xfadevalue = w->xfadevalue; float xfadelength = w->xfadelength; float xfadespeed, xfadestep, readindex, outputsample; if(pitchfactor > 0.5) xfadespeed = pitchfactor; else xfadespeed = 1 - pitchfactor; xfadestep = xfadespeed / xfadelength; while(n--) { if(readlag > period) // check if read pointer may jump forward... { if(xfadevalue <= 0) // ...but do not interrupt crossfade { jump = period; // jump forward while((jump * 2) < readlag) jump *= 2; // use available space readlag -= jump; // reduce read pointer lag xfadevalue = 1; // start crossfade xfadelength = period - 1; xfadestep = xfadespeed / xfadelength; } } readindex = refindex - readlag; outputsample = read_sample(w, readindex); if(xfadevalue > 0) { outputsample *= (1 - xfadevalue); // fadein outputsample += read_sample(w, readindex - jump) * xfadevalue; // fadeout xfadevalue -= xfadestep; } *out++ = outputsample; refindex += 1; readlag += readlagstep; } w->jump = jump; // state variables w->readlag = readlag; w->xfadevalue = xfadevalue; w->xfadelength = xfadelength; } /* Function pitchup() for pitch factors above 1 is more complicated than pitchdown(). The read pointer increments faster than the write pointer and a backward jump must happen in time, reckoning with the crossfade region. The read pointer backward jump length is always one period. In order to minimize the area of signal duplicates, crossfade length is aimed at [period / pitchfactor]. This leads to a crossfade speed of [pitchfactor * pitchfactor]. Some samples for the fade out (but not all of them) must already be in the buffer, otherwise we will run out of input samples before the crossfade is completed. The ratio of past samples and future samples for a crossfade of any length is as follows: past samples: xfadelength * (1 - 1 / pitchfactor) future samples: xfadelength * (1 / pitchfactor) For example in the case of pitch factor 1.5 this would be: past samples: xfadelength * (1 - 1 / 1.5) = xfadelength * 1 / 3 future samples: xfadelength * (1 / 1.5) = xfadelength * 2 / 3 In the case of pitch factor 4 this would be: past samples: xfadelength * (1 - 1 / 4) = xfadelength * 3 / 4 future samples: xfadelength * (1 / 4) = xfadelength * 1 / 4 The read pointer lag must therefore preserve a minimum dependent on pitch factor. The minimum is called 'limit' here: limit = period * (pitchfactor - 1) / pitchfactor * pitchfactor Components of this expression are combined to reuse them in operations, while (pitchfactor - 1) is changed to (pitchfactor - 0.99) to avoid numerical resolution issues for pitch factors slightly above 1: xfadespeed = pitchfactor * pitchfactor limitfactor = (pitchfactor - 0.99) / xfadespeed limit = period * limitfactor When read lag is smaller than this limit, the read pointer must preferably jump backward, unless a previous crossfade is not yet completed. Crossfades must preferably be completed, unless the read pointer lag becomes smaller than zero. With fluctuating period lengths and pitch factors, the readpointer lag limit may change from one input block to the next in such a way that the actual lag is suddenly much smaller than the limit, and the intended crossfade length can not be applied. Therefore the crossfade length is simply calculated from the available amount of samples for all cases, like so: xfadelength = readlag / limitfactor For most occurrences, this will amount to a crossfade length reduced to [period / pitchfactor] in the output for pitch factors above 1, while in some cases it will be considerably shorter. Fortunately, an incidental aberration of the intended crossfade length hardly ever creates an audible artifact. The reason to specify preferred crossfade length according to pitch factor is to minimize the impression of echoes without sacrificing too much of the signal content. The readpointer jump length remains one period in any case. Sometimes, the input signal periodicity may decrease substantially between one signal block and the next. In such cases it may be possible for the read pointer to jump forward and reduce latency. For every signal block, a check on this possibility is done. A previous crossfade must be completed before a forward jump is allowed. */ static void pitchup(tSOLAD* const w, float *out) { int n = w->blocksize; float refindex = (float)(w->timeindex + LOOPSIZE); // no negative values float pitchfactor = w->pitchfactor; float period = w->period; float readlag = w->readlag; float jump = w->jump; float xfadevalue = w->xfadevalue; float xfadelength = w->xfadelength; float readlagstep = pitchfactor - 1; float xfadespeed = pitchfactor * pitchfactor; float xfadestep = xfadespeed / xfadelength; float limitfactor = (pitchfactor - (float)0.99) / xfadespeed; float limit = period * limitfactor; float readindex, outputsample; if((readlag > (period + 2 * limit)) & (xfadevalue < 0)) { jump = period; // jump forward while((jump * 2) < (readlag - 2 * limit)) jump *= 2; // use available space readlag -= jump; // reduce read pointer lag xfadevalue = 1; // start crossfade xfadelength = period - 1; xfadestep = xfadespeed / xfadelength; } while(n--) { if(readlag < limit) // check if read pointer should jump backward... { if((xfadevalue < 0) | (readlag < 0)) // ...but try not to interrupt crossfade { xfadelength = readlag / limitfactor; if(xfadelength < 1) xfadelength = 1; xfadestep = xfadespeed / xfadelength; jump = -period; // jump backward readlag += period; // increase read pointer lag xfadevalue = 1; // start crossfade } } readindex = refindex - readlag; outputsample = read_sample(w, readindex); if(xfadevalue > 0) { outputsample *= (1 - xfadevalue); outputsample += read_sample(w, readindex - jump) * xfadevalue; xfadevalue -= xfadestep; } *out++ = outputsample; refindex += 1; readlag -= readlagstep; } w->readlag = readlag; // state variables w->jump = jump; w->xfadelength = xfadelength; w->xfadevalue = xfadevalue; } // read one sample from delay buffer, with linear interpolation static inline float read_sample(tSOLAD* const w, float floatindex) { int index = (int)floatindex; float fraction = floatindex - (float)index; float *buf = w->delaybuf; index &= LOOPMASK; return (buf[index] + (fraction * (buf[index+1] - buf[index]))); } static void solad_init(tSOLAD* const w) { w->timeindex = 0; w->xfadevalue = -1; w->period = INITPERIOD; w->readlag = INITPERIOD; w->blocksize = INITPERIOD; } /******************************************************************************/ /***************************** private procedures *****************************/ /******************************************************************************/ #define REALFFT mayer_realfft #define REALIFFT mayer_realifft static void snac_analyzeframe(tSNAC* const s); static void snac_autocorrelation(tSNAC* const s); static void snac_normalize(tSNAC* const s); static void snac_pickpeak(tSNAC* const s); static void snac_periodandfidelity(tSNAC* const s); static void snac_biasbuf(tSNAC* const s); static float snac_spectralpeak(tSNAC* const s, float periodlength); /******************************************************************************/ /******************************** constructor, destructor *********************/ /******************************************************************************/ void tSNAC_init(tSNAC* const s, int overlaparg) { s->biasfactor = DEFBIAS; s->timeindex = 0; s->periodindex = 0; s->periodlength = 0.; s->fidelity = 0.; s->minrms = DEFMINRMS; s->framesize = SNAC_FRAME_SIZE; snac_biasbuf(s); tSNAC_setOverlap(s, overlaparg); } void tSNAC_free(tSNAC* const s) { leaf_free(s); } /******************************************************************************/ /************************** public access functions****************************/ /******************************************************************************/ void tSNAC_ioSamples(tSNAC* const s, float *in, float *out, int size) { int timeindex = s->timeindex; int mask = s->framesize - 1; int outindex = 0; float *inputbuf = s->inputbuf; float *processbuf = s->processbuf; // call analysis function when it is time if(!(timeindex & (s->framesize / s->overlap - 1))) snac_analyzeframe(s); while(size--) { inputbuf[timeindex] = *in++; out[outindex++] = processbuf[timeindex++]; timeindex &= mask; } s->timeindex = timeindex; } void tSNAC_setOverlap(tSNAC* const s, int lap) { if(!((lap==1)|(lap==2)|(lap==4)|(lap==8))) lap = DEFOVERLAP; s->overlap = lap; } void tSNAC_setBias(tSNAC* const s, float bias) { if(bias > 1.) bias = 1.; if(bias < 0.) bias = 0.; s->biasfactor = bias; snac_biasbuf(s); return; } void tSNAC_setMinRMS(tSNAC* const s, float rms) { if(rms > 1.) rms = 1.; if(rms < 0.) rms = 0.; s->minrms = rms; return; } float tSNAC_getPeriod(tSNAC* const s) { return(s->periodlength); } float tSNAC_getFidelity(tSNAC* const s) { return(s->fidelity); } /******************************************************************************/ /***************************** private procedures *****************************/ /******************************************************************************/ // main analysis function static void snac_analyzeframe(tSNAC* const s) { int n, tindex = s->timeindex; int framesize = s->framesize; int mask = framesize - 1; float norm = 1. / sqrt((float)(framesize * 2)); float *inputbuf = s->inputbuf; float *processbuf = s->processbuf; // copy input to processing buffers for(n=0; n<framesize; n++) { processbuf[n] = inputbuf[tindex] * norm; tindex++; tindex &= mask; } // zeropadding for(n=framesize; n<(framesize<<1); n++) processbuf[n] = 0.; // call analysis procedures snac_autocorrelation(s); snac_normalize(s); snac_pickpeak(s); snac_periodandfidelity(s); } static void snac_autocorrelation(tSNAC* const s) { int n, m; int framesize = s->framesize; int fftsize = framesize * 2; float *processbuf = s->processbuf; float *spectrumbuf = s->spectrumbuf; REALFFT(fftsize, processbuf); // compute power spectrum processbuf[0] *= processbuf[0]; // DC processbuf[framesize] *= processbuf[framesize]; // Nyquist for(n=1; n<framesize; n++) { processbuf[n] = processbuf[n] * processbuf[n] + processbuf[fftsize-n] * processbuf[fftsize-n]; // imag coefficients appear reversed processbuf[fftsize-n] = 0.; } // store power spectrum up to SR/4 for possible later use for(m=0; m<(framesize>>1); m++) { spectrumbuf[m] = processbuf[m]; } // transform power spectrum to autocorrelation function REALIFFT(fftsize, processbuf); return; } static void snac_normalize(tSNAC* const s) { int framesize = s->framesize; int framesizeplustimeindex = s->framesize + s->timeindex; int timeindexminusone = s->timeindex - 1; int n, m; int mask = framesize - 1; int seek = framesize * SEEK; float *inputbuf = s->inputbuf; float *processbuf= s->processbuf; float signal1, signal2; // minimum RMS implemented as minimum autocorrelation at index 0 // functionally equivalent to white noise floor float rms = s->minrms / sqrt(1.0f / (float)framesize); float minrzero = rms * rms; float rzero = processbuf[0]; if(rzero < minrzero) rzero = minrzero; double normintegral = (double)rzero * 2.; // normalize biased autocorrelation function // inputbuf is circular buffer: timeindex may be non-zero when overlap > 1 processbuf[0] = 1; for(n=1, m=s->timeindex+1; n<seek; n++, m++) { signal1 = inputbuf[(n + timeindexminusone)&mask]; signal2 = inputbuf[(framesizeplustimeindex - n)&mask]; normintegral -= (double)(signal1 * signal1 + signal2 * signal2); processbuf[n] /= (float)normintegral * 0.5f; } // flush instable function tail for(n = seek; n<framesize; n++) processbuf[n] = 0.; return; } static void snac_periodandfidelity(tSNAC* const s) { float periodlength; if(s->periodindex) { periodlength = (float)s->periodindex + interpolate3phase(s->processbuf, s->periodindex); if(periodlength < 8) periodlength = snac_spectralpeak(s, periodlength); s->periodlength = periodlength; s->fidelity = interpolate3max(s->processbuf, s->periodindex); } return; } // select the peak which most probably represents period length static void snac_pickpeak(tSNAC* const s) { int n, peakindex=0; int seek = s->framesize * SEEK; float *processbuf= s->processbuf; float maxvalue = 0.; float biasedpeak; float *biasbuf = s->biasbuf; // skip main lobe for(n=1; n<seek; n++) { if(processbuf[n] < 0.) break; } // find interpolated / biased maximum in SNAC function // interpolation finds the 'real maximum' // biasing favours the first candidate for(; n<seek-1; n++) { if(processbuf[n] >= processbuf[n-1]) { if(processbuf[n] > processbuf[n+1]) // we have a local peak { biasedpeak = interpolate3max(processbuf, n) * biasbuf[n]; if(biasedpeak > maxvalue) { maxvalue = biasedpeak; peakindex = n; } } } } s->periodindex = peakindex; return; } // verify period length via frequency domain (up till SR/4) // frequency domain is more precise than lag domain for period lengths < 8 // argument 'periodlength' is initial estimation from autocorrelation static float snac_spectralpeak(tSNAC* const s, float periodlength) { if(periodlength < 4.) return periodlength; float max = 0.; int n, startbin, stopbin, peakbin = 0; int spectrumsize = s->framesize>>1; float *spectrumbuf = s->spectrumbuf; float peaklocation = (float)(s->framesize * 2.) / periodlength; startbin = (int)(peaklocation * 0.8 + 0.5); if(startbin < 1) startbin = 1; stopbin = (int)(peaklocation * 1.25 + 0.5); if(stopbin >= spectrumsize - 1) stopbin = spectrumsize - 1; for(n=startbin; n<stopbin; n++) { if(spectrumbuf[n] >= spectrumbuf[n-1]) { if(spectrumbuf[n] > spectrumbuf[n+1]) { if(spectrumbuf[n] > max) { max = spectrumbuf[n]; peakbin = n; } } } } // calculate amplitudes in peak region for(n=(peakbin-1); n<(peakbin+2); n++) { spectrumbuf[n] = sqrt(spectrumbuf[n]); } peaklocation = (float)peakbin + interpolate3phase(spectrumbuf, peakbin); periodlength = (float)(s->framesize * 2.0f) / peaklocation; return periodlength; } // modified logarithmic bias function static void snac_biasbuf(tSNAC* const s) { int n; int maxperiod = (int)(s->framesize * (float)SEEK); float bias = s->biasfactor / log((float)(maxperiod - 4)); float *biasbuf = s->biasbuf; for(n=0; n<5; n++) // periods < 5 samples can't be tracked { biasbuf[n] = 0.; } for(n=5; n<maxperiod; n++) { biasbuf[n] = 1.0f - (float)log(n - 4) * bias; } } /********Private function prototypes**********/ static void atkdtk_init(tAtkDtk* const a, int blocksize, int atk, int rel); static void atkdtk_envelope(tAtkDtk* const a, float *in); /********Constructor/Destructor***************/ void tAtkDtk_init(tAtkDtk* const a, int blocksize) { atkdtk_init(a, blocksize, DEFATTACK, DEFRELEASE); } void tAtkDtk_init_expanded(tAtkDtk* const a, int blocksize, int atk, int rel) { atkdtk_init(a, blocksize, atk, rel); } void tAtkDtk_free(tAtkDtk *a) { leaf_free(a); } /*******Public Functions***********/ void tAtkDtk_setBlocksize(tAtkDtk* const a, int size) { if(!((size==64)|(size==128)|(size==256)|(size==512)|(size==1024)|(size==2048))) size = DEFBLOCKSIZE; a->blocksize = size; return; } void tAtkDtk_setSamplerate(tAtkDtk* const a, int inRate) { a->samplerate = inRate; //Reset atk and rel to recalculate coeff tAtkDtk_setAtk(a, a->atk); tAtkDtk_setRel(a, a->rel); return; } void tAtkDtk_setThreshold(tAtkDtk* const a, float thres) { a->threshold = thres; return; } void tAtkDtk_setAtk(tAtkDtk* const a, int inAtk) { a->atk = inAtk; a->atk_coeff = pow(0.01, 1.0/(a->atk * a->samplerate * 0.001)); return; } void tAtkDtk_setRel(tAtkDtk* const a, int inRel) { a->rel = inRel; a->rel_coeff = pow(0.01, 1.0/(a->rel * a->samplerate * 0.001)); return; } int tAtkDtk_detect(tAtkDtk* const a, float *in) { int result; atkdtk_envelope(a, in); if(a->env >= a->prevAmp*2) //2 times greater = 6dB increase result = 1; else result = 0; a->prevAmp = a->env; return result; } /*******Private Functions**********/ static void atkdtk_init(tAtkDtk* const a, int blocksize, int atk, int rel) { a->env = 0; a->blocksize = blocksize; a->threshold = DEFTHRESHOLD; a->samplerate = leaf.sampleRate; a->prevAmp = 0; a->env = 0; tAtkDtk_setAtk(a, atk); tAtkDtk_setRel(a, rel); } static void atkdtk_envelope(tAtkDtk* const a, float *in) { int i = 0; float tmp; for(i = 0; i < a->blocksize; ++i){ tmp = fastabs(in[i]); if(tmp > a->env) a->env = a->atk_coeff * (a->env - tmp) + tmp; else a->env = a->rel_coeff * (a->env - tmp) + tmp; } }