void ofxFFT::calc() {
//Generate a split complex vector from the real data
vDSP_ctoz((COMPLEX *)input, 2, &mDspSplitComplex, 1, mFFTLength);
//Take the fft and scale appropriately
vDSP_fft_zrip(mSpectrumAnalysis, &mDspSplitComplex, 1, log2n, FFT_FORWARD);
// vDSP_fft_zrip(mSpectrumAnalysis, &mDspSplitComplex, 1, log2n, FFT_INVERSE);
vDSP_vsmul(mDspSplitComplex.realp, 1, &mFFTNormFactor, mDspSplitComplex.realp, 1, mFFTLength);
vDSP_vsmul(mDspSplitComplex.imagp, 1, &mFFTNormFactor, mDspSplitComplex.imagp, 1, mFFTLength);
// /* The output signal is now in a split real form. Use the function
// * vDSP_ztoc to get a split real vector. */
// vDSP_ztoc(&mDspSplitComplex, 1, (COMPLEX *) output, 2, mFFTLength);
//
//Zero out the nyquist value
mDspSplitComplex.imagp[0] = 0.0;
//Convert the fft data to dB
vDSP_zvmags(&mDspSplitComplex, 1, amplitude, 1, mFFTLength);
//In order to avoid taking log10 of zero, an adjusting factor is added in to make the minimum value equal -128dB
float mAdjust0DB = ADJUST_0_DB;
vDSP_vsadd(amplitude, 1, &mAdjust0DB, power, 1, mFFTLength);
float one = 1;
vDSP_vdbcon(power, 1, &one, power, 1, mFFTLength, 0);
}
/*******************************************************************************
VectorScalarAdd */
Error_t
VectorScalarAdd(float *dest,
const float *in1,
const float scalar,
unsigned length)
{
#ifdef __APPLE__
// Use the Accelerate framework if we have it
vDSP_vsadd(in1, 1, &scalar, dest, 1, length);
#else
// Otherwise do it manually
unsigned i;
const unsigned end = 4 * (length / 4);
for (i = 0; i < end; i+=4)
{
dest[i] = in1[i] + scalar;
dest[i + 1] = in1[i + 1] + scalar;
dest[i + 2] = in1[i + 2] + scalar;
dest[i + 3] = in1[i + 3] + scalar;
}
for (i = end; i < length; ++i)
{
dest[i] = in1[i] + scalar;
}
#endif
return NOERR;
}
void CsoundObject_writePVSReadPath(CsoundObject *self,
size_t *segmentStartFrames,
size_t triangleFilterBandsCount,
Matrix32 *warpPaths)
{
Float32 frameLengthInSeconds = 1./44100. * 256.;
for (size_t i = 0; i < triangleFilterBandsCount; ++i) {
Float32 startFrameInSeconds = (Float32)segmentStartFrames[i] * frameLengthInSeconds;
Float32 *tablePointer;
csoundGetTable(self->csound, &tablePointer, (SInt32)(warpPathTableBaseNumber + i));
vDSP_vsadd(Matrix_getRow(warpPaths, i), 1, &startFrameInSeconds, tablePointer, 1, self->analysisSegmentFramesCount);
}
}
void FFTHelper::ComputeFFT(Float32* inAudioData, Float32* outFFTData)
{
if (inAudioData == NULL || outFFTData == NULL) return;
//Generate a split complex vector from the real data
vDSP_ctoz((COMPLEX *)inAudioData, 2, &mDspSplitComplex, 1, mFFTLength);
//Take the fft and scale appropriately
vDSP_fft_zrip(mSpectrumAnalysis, &mDspSplitComplex, 1, mLog2N, kFFTDirection_Forward);
vDSP_vsmul(mDspSplitComplex.realp, 1, &mFFTNormFactor, mDspSplitComplex.realp, 1, mFFTLength);
vDSP_vsmul(mDspSplitComplex.imagp, 1, &mFFTNormFactor, mDspSplitComplex.imagp, 1, mFFTLength);
//Zero out the nyquist value
mDspSplitComplex.imagp[0] = 0.0;
//Convert the fft data to dB
vDSP_zvmags(&mDspSplitComplex, 1, outFFTData, 1, mFFTLength);
//In order to avoid taking log10 of zero, an adjusting factor is added in to make the minimum value equal -128dB
vDSP_vsadd(outFFTData, 1, &kAdjust0DB, outFFTData, 1, mFFTLength);
Float32 one = 1;
vDSP_vdbcon(outFFTData, 1, &one, outFFTData, 1, mFFTLength, 0);
Float32 max = -100;
int index = -1;
for(unsigned long i = 0; i < mFFTLength; i++){
if(outFFTData[i] > max){
max = outFFTData[i];
index = i;
}
}
if(max > -40){ // Filter out anything else, as it is unlikely to be the microwave beep
recentMaxIndex = index;
//if(index == 181){ // We found the microwave beep
//printf("%d %f\n", index, max);
}else{
recentMaxIndex = 0;
}
}
void FFTHelper::ComputeFFT(Float32* inAudioData, Float32* outFFTData)
{
if (inAudioData == NULL || outFFTData == NULL) return;
//Generate a split complex vector from the real data
vDSP_ctoz((COMPLEX *)inAudioData, 2, &mDspSplitComplex, 1, mFFTLength);
//Take the fft and scale appropriately
vDSP_fft_zrip(mSpectrumAnalysis, &mDspSplitComplex, 1, mLog2N, kFFTDirection_Forward);
vDSP_vsmul(mDspSplitComplex.realp, 1, &mFFTNormFactor, mDspSplitComplex.realp, 1, mFFTLength);
vDSP_vsmul(mDspSplitComplex.imagp, 1, &mFFTNormFactor, mDspSplitComplex.imagp, 1, mFFTLength);
//Zero out the nyquist value
mDspSplitComplex.imagp[0] = 0.0;
//Convert the fft data to dB
vDSP_zvmags(&mDspSplitComplex, 1, outFFTData, 1, mFFTLength);
//In order to avoid taking log10 of zero, an adjusting factor is added in to make the minimum value equal -128dB
vDSP_vsadd(outFFTData, 1, &kAdjust0DB, outFFTData, 1, mFFTLength);
Float32 one = 1;
vDSP_vdbcon(outFFTData, 1, &one, outFFTData, 1, mFFTLength, 0);
}
void DspTableRead4::processDspWithIndex(int fromIndex, int toIndex) {
if (table != NULL) { // ensure that there is a table to read from!
int bufferLength;
float *buffer = table->getBuffer(&bufferLength);
if (ArrayArithmetic::hasAccelerate) {
#if __APPLE__
//float zero = 0.0f;
//float bufferLengthFloat = (float) (bufferLength-2);
//vDSP_vclip(inputBuffer+startSampleIndex, 1, &zero, &bufferLengthFloat,
// inputBuffer+startSampleIndex, 1, endSampleIndex-startSampleIndex);
// NOTE(mhroth): is isn't clear what the clipping behaviour of vDSP_vlint is, but I
// *think* that it is doing the right thing (i.e., clipping OOB indicies)
vDSP_vsadd(dspBufferAtInlet0+fromIndex, 1, &offset, dspBufferAtOutlet0+fromIndex, 1,
toIndex-fromIndex);
vDSP_vlint(buffer, dspBufferAtInlet0+fromIndex, 1, dspBufferAtOutlet0+fromIndex, 1,
toIndex-fromIndex, bufferLength);
#endif
} else {
int maxIndex = bufferLength-1;
for (int i = fromIndex; i < toIndex; i++) {
int x = (int) (dspBufferAtInlet0[i] + offset);
if (x <= 0) {
dspBufferAtOutlet0[i] = buffer[0];
} else if (x >= maxIndex) {
dspBufferAtOutlet0[i] = buffer[maxIndex];
} else {
// 2-point linear interpolation (basic and fast)
float dx = dspBufferAtInlet0[i] - ((float) x);
float y0 = buffer[x];
float y1 = buffer[x+1];
float slope = (y1 - y0);
dspBufferAtOutlet0[i] = (slope * dx) + y0;
}
}
}
}
}
void sub( const float *array, float scalar, float *result, size_t length )
{
scalar *= -1;
vDSP_vsadd( const_cast<float *>( array ), 1, &scalar, result, 1, length );
}
TriangleFilterBank32 *TriangleFilterBank32_new(size_t filterCount, size_t magnitudeFrameSize, size_t samplerate)
{
TriangleFilterBank32 *self = calloc(1, sizeof(TriangleFilterBank32));
self->filterCount = filterCount;
self->filterFrequencyCount = filterCount + 2;
self->magnitudeFrameSize = magnitudeFrameSize;
self->samplerate = samplerate;
self->filterFrequencies = calloc(self->filterFrequencyCount, sizeof(Float32));
self->filterBank = calloc(self->magnitudeFrameSize * self->filterCount, sizeof(Float32));
Float32 *filterTemp = calloc(self->magnitudeFrameSize * self->filterCount, sizeof(Float32));
Float32 one = 2;
Float32 zero = 0;
vDSP_vgen(&zero, &one, self->filterFrequencies, 1, self->filterFrequencyCount);
for (size_t i = 0; i < self->filterFrequencyCount; ++i) {
self->filterFrequencies[i] = powf(10, self->filterFrequencies[i]);
}
Float32 minusFirstElement = -self->filterFrequencies[0];
vDSP_vsadd(self->filterFrequencies, 1, &minusFirstElement, self->filterFrequencies, 1, self->filterFrequencyCount);
Float32 maximum = self->filterFrequencies[self->filterFrequencyCount - 1];
vDSP_vsdiv(self->filterFrequencies, 1, &maximum, self->filterFrequencies, 1, self->filterFrequencyCount);
Float32 nyquist = self->samplerate / 2;
vDSP_vsmul(self->filterFrequencies, 1, &nyquist, self->filterFrequencies, 1, self->filterFrequencyCount);
Float32 *magnitudeFrequencies = calloc(self->magnitudeFrameSize, sizeof(Float32));
vDSP_vgen(&zero, &nyquist, magnitudeFrequencies, 1, self->magnitudeFrameSize);
Float32 *lower = calloc(self->filterCount, sizeof(Float32));
Float32 *center = calloc(self->filterCount, sizeof(Float32));
Float32 *upper = calloc(self->filterCount, sizeof(Float32));
Float32 *temp1 = (Float32 *)calloc(self->magnitudeFrameSize, sizeof(Float32));
Float32 *temp2 = (Float32 *)calloc(self->magnitudeFrameSize, sizeof(Float32));
cblas_scopy((SInt32)self->filterCount, &self->filterFrequencies[0], 1, lower, 1);
cblas_scopy((SInt32)self->filterCount, &self->filterFrequencies[1], 1, center, 1);
cblas_scopy((SInt32)self->filterCount, &self->filterFrequencies[2], 1, upper, 1);
for (size_t i = 0; i < self->filterCount; ++i) {
Float32 negateLowerValue = -lower[i];
vDSP_vsadd(magnitudeFrequencies, 1, &negateLowerValue, temp1, 1, self->magnitudeFrameSize);
Float32 divider = center[i] - lower[i];
vDSP_vsdiv(temp1, 1, ÷r, temp1, 1, self->magnitudeFrameSize);
for (size_t j = 0; j < self->magnitudeFrameSize; ++j) {
if (!(magnitudeFrequencies[j] > lower[i] && magnitudeFrequencies[j] <= center[i])) {
temp1[j] = 0;
}
}
Float32 minusOne = -1;
vDSP_vsmul(magnitudeFrequencies, 1, &minusOne, temp2, 1, self->magnitudeFrameSize);
vDSP_vsadd(temp2, 1, &upper[i], temp2, 1, self->magnitudeFrameSize);
divider = upper[i] - center[i];
vDSP_vsdiv(temp2, 1, ÷r, temp2, 1, self->magnitudeFrameSize);
for (size_t j = 0; j < self->magnitudeFrameSize; ++j) {
if (!(magnitudeFrequencies[j] > center[i] && magnitudeFrequencies[j] <= upper[i])) {
temp2[j] = 0;
}
}
vDSP_vadd(temp1, 1, temp2, 1, &filterTemp[i * self->magnitudeFrameSize], 1, self->magnitudeFrameSize);
}
vDSP_mtrans(filterTemp, 1, self->filterBank, 1, self->magnitudeFrameSize, self->filterCount);
free(lower);
free(center);
free(upper);
free(temp1);
free(temp2);
free(magnitudeFrequencies);
free(filterTemp);
return self;
}
Boolean FFTBufferManager::ComputeFFT(int32_t *outFFTData)
{
if (HasNewAudioData())
{
// for(int i=0;i<mFFTLength;i++)
// {
// if(mAudioBuffer[i]>0.15)
// printf("%f\n",mAudioBuffer[i]);
// }
//Generate a split complex vector from the real data
# define NOISE_FILTER 0.01;
for(int i=0;i<4096;i++)
{
//DENOISE
if(mAudioBuffer[i]>0)
{
mAudioBuffer[i] -= NOISE_FILTER;
if(mAudioBuffer[i] < 0)
{
mAudioBuffer[i] = 0;
}
}
else if(mAudioBuffer[i]<0)
{
mAudioBuffer[i] += NOISE_FILTER;
if(mAudioBuffer[i]>0)
{
mAudioBuffer[i] = 0;
}
}
else
{
mAudioBuffer[i] = 0;
}
}
vDSP_ctoz((COMPLEX *)mAudioBuffer, 2, &mDspSplitComplex, 1, mFFTLength);
//Take the fft and scale appropriately
vDSP_fft_zrip(mSpectrumAnalysis, &mDspSplitComplex, 1, mLog2N, kFFTDirection_Forward);
vDSP_vsmul(mDspSplitComplex.realp, 1, &mFFTNormFactor, mDspSplitComplex.realp, 1, mFFTLength);
vDSP_vsmul(mDspSplitComplex.imagp, 1, &mFFTNormFactor, mDspSplitComplex.imagp, 1, mFFTLength);
//Zero out the nyquist value
mDspSplitComplex.imagp[0] = 0.0;
//Convert the fft data to dB
Float32 tmpData[mFFTLength];
vDSP_zvmags(&mDspSplitComplex, 1, tmpData, 1, mFFTLength);
//In order to avoid taking log10 of zero, an adjusting factor is added in to make the minimum value equal -128dB
vDSP_vsadd(tmpData, 1, &mAdjust0DB, tmpData, 1, mFFTLength);
Float32 one = 1;
vDSP_vdbcon(tmpData, 1, &one, tmpData, 1, mFFTLength, 0);
//Convert floating point data to integer (Q7.24)
vDSP_vsmul(tmpData, 1, &m24BitFracScale, tmpData, 1, mFFTLength);
for(UInt32 i=0; i<mFFTLength; ++i)
outFFTData[i] = (SInt32) tmpData[i];
for(int i=0;i<mFFTLength/4;i++)
{
printf("%i:::%i\n",i,outFFTData[i]/16777216+120);
}
OSAtomicDecrement32Barrier(&mHasAudioData);
OSAtomicIncrement32Barrier(&mNeedsAudioData);
mAudioBufferCurrentIndex = 0;
return true;
}
else if (mNeedsAudioData == 0)
OSAtomicIncrement32Barrier(&mNeedsAudioData);
return false;
}
bool ofxAudioUnitFftNode::getAmplitude(std::vector<float> &outAmplitude)
{
getSamplesFromChannel(_sampleBuffer, 0);
// return empty if we don't have enough samples yet
if(_sampleBuffer.size() < _N) {
outAmplitude.clear();
return false;
}
// normalize input waveform
if(_outputSettings.normalizeInput) {
float timeDomainMax;
vDSP_maxv(&_sampleBuffer[0], 1, &timeDomainMax, _N);
vDSP_vsdiv(&_sampleBuffer[0], 1, &timeDomainMax, &_sampleBuffer[0], 1, _N);
}
PerformFFT(&_sampleBuffer[0], _window, _fftData, _fftSetup, _N);
// get amplitude
vDSP_zvmags(&_fftData, 1, _fftData.realp, 1, _N/2);
// normalize magnitudes
float two = 2.0;
vDSP_vsdiv(_fftData.realp, 1, &two, _fftData.realp, 1, _N/2);
// scale output according to requested settings
if(_outputSettings.scale == OFXAU_SCALE_LOG10) {
for(int i = 0; i < (_N / 2); i++) {
_fftData.realp[i] = log10f(_fftData.realp[i] + 1);
}
} else if(_outputSettings.scale == OFXAU_SCALE_DECIBEL) {
float ref = 1.0;
vDSP_vdbcon(_fftData.realp, 1, &ref, _fftData.realp, 1, _N / 2, 1);
float dbCorrectionFactor = 0;
switch (_outputSettings.window) {
case OFXAU_WINDOW_HAMMING:
dbCorrectionFactor = DB_CORRECTION_HAMMING;
break;
case OFXAU_WINDOW_HANNING:
dbCorrectionFactor = DB_CORRECTION_HAMMING;
break;
case OFXAU_WINDOW_BLACKMAN:
dbCorrectionFactor = DB_CORRECTION_HAMMING;
break;
}
vDSP_vsadd(_fftData.realp, 1, &dbCorrectionFactor, _fftData.realp, 1, _N / 2);
}
// restrict minimum to 0
if(_outputSettings.clampMinToZero) {
float min = 0.0;
float max = INFINITY;
vDSP_vclip(_fftData.realp, 1, &min, &max, _fftData.realp, 1, _N / 2);
}
// normalize output between 0 and 1
if(_outputSettings.normalizeOutput) {
float max;
vDSP_maxv(_fftData.realp, 1, &max, _N / 2);
if(max > 0) {
vDSP_vsdiv(_fftData.realp, 1, &max, _fftData.realp, 1, _N / 2);
}
}
outAmplitude.assign(_fftData.realp, _fftData.realp + _N/2);
return true;
}
