int main()
{
// Set the global sample rate before creating class instances.
Stk::setSampleRate( 44100.0 );
int nFrames = 100000;
FileLoop input;
FileWvOut output;
try {
// Load the sine wave file.
input.openFile( "rawwaves/sinewave.raw", true );
// Open a 16-bit, one-channel WAV formatted output file
output.openFile( "hellosine.wav", 1, FileWrite::FILE_WAV, Stk::STK_SINT16 );
}
catch ( StkError & ) {
exit( 1 );
}
input.setFrequency( 440.0 );
// Option 1: Use StkFrames
/*
StkFrames frames( nFrames, 1 );
try {
output.tick( input.tick( frames ) );
}
catch ( StkError & ) {
exit( 1 );
}
*/
// Option 2: Single-sample computations
for ( int i=0; i<nFrames; i++ ) {
try {
output.tick( input.tick() );
}
catch ( StkError & ) {
exit( 1 );
}
}
return 0;
}
int main(int argc, char *argv[])
{
double velSource;
double velReciver;
double distX;
double distY;
double velRelativeX;
double totalTime;
unsigned long numSamples;
if ( argc < 7 || argc > 8 ) {
std::cout<<"\n\n\n****Doppler shift example*****\n\n";
std::cout<<"Will produce output assuming source is travelling towards\n";
std::cout<<"and then past receiver in straight line\n\n";
std::cout<<"Will start simulation at d0=-D and end at d1=+D\n";
std::cout<<"Simulation time = 2D/(relative velocity)\n";
std::cout<<"input will be looped for duration of simulation\n\n";
std::cout<< "usage: doppler [spd src (m/s)] [spd recv (m/s)] [distX (m)] [distY (m)] [input file (wav)] [output file (wav)] [0/1: no mix/mix]\n";
std::cout<< "NOTE: if source and receiver are diverging, fix simulation duration to 5s\n";
return 0;
}
else {
std::cout<<"INPUT OK...\n parsed values:";
velSource = atof(argv[1]);
velReciver = atof(argv[2]);
distX = atof(argv[3]);
distY = atof(argv[4]);
//visual check to make sure input was parsed properly
std::cout<<"velSource=" << velSource << " velReciver=" << velReciver << " distX=" << distX << " distY=" <<distY << " if=" << argv[5] << "\n\n";
velRelativeX = velSource-velReciver;
double distAbs = sqrt(distX*distX + distY*distY);
totalTime = 2*distX/velRelativeX; //total simulation time, if not diverging/constant
//do some basic input checking:
if ((velSource - velReciver) <= 0) {
//this is for the "not so interesting" case where
//the source and receiver will never "cross"
// will just have a boring, constant frequency...
//so we force a fixed simulation time here.
std::cout<<"DIVERGING! clamping time to 5 seconds\n";
totalTime = 5.0;
}
numSamples = totalTime * SAMPLE_RATE;
std::cout<<"relative speedX = " << velRelativeX<<"; simulation time = " << totalTime<< "s; numSamples = "<< numSamples<<"\n";
}
Stk::setSampleRate( SAMPLE_RATE );
//file i/o
FileWvOut outputFile;
FileLoop inputLoop;
try {
outputFile.openFile( argv[6], 1, FileWrite::FILE_WAV, Stk::STK_SINT16 );
inputLoop.openFile(argv[5]);
}
catch (StkError &) {
exit( 1 );
}
//the doppler effect's frequency shift is implemented using a delay line whose write
// pointer is incremented sample by sample and read using a variable delay.
DelayL delay(1.0, DELAY_SIZE);
//initialize: since read pointer can increment faster than write pointer, we should
// fill up the delay line with initial values for continuity's sake...
// since our input is a loop, it doesn't matter too much where the "starting" point
// in the audio sample is...
for (int i=0; i<DELAY_SIZE; i++) {
delay.tick(inputLoop.tick());
}
stk::StkFrames frames( numSamples, 1);
double readPtr = 0;
unsigned int writePtr = 0;
double g=0;
// set initial locations
double xSrcCurr = 0, xSrcPrev = 0;
double xRcvCurr = distX, xRcvPrev = distX;
double dT = totalTime/numSamples;
for (int i=0; i<numSamples; i++) {
//basic kinematic model for estimating current relative speed:
double timeElapsed = (double)totalTime*i/numSamples;
xSrcCurr+=dT*velSource;
xRcvCurr+=dT*velReciver;
//distY is constant!
//velRelative = ( sqrt((xSrcCurr-xSrcPrev)*(xSrcCurr-xSrcPrev)+distY*distY) +
// sqrt((xRcvPrev-xRcvCurr)*(-xRcvCurr)+distY*distY) ) / dT;
//double velRelative = ( (xRcvCurr-xSrcCurr) - (xRcvPrev-xSrcPrev) ) / dT;
//bit of a hack to get the right behaviour, as my quickly cobbled together model
// was bit sloppy when it came to managing the sign
double velRelative = -fabs( sqrt((xRcvCurr-xSrcCurr)*(xRcvCurr-xSrcCurr)+distY*distY)
- sqrt((xRcvPrev-xSrcPrev)*(xRcvPrev-xSrcPrev)+distY*distY) ) / dT;
if (xRcvCurr < xSrcCurr)
velRelative = -velRelative;
//std::cout <<xSrcCurr<<" "<<xRcvCurr<< " "<< velRelative <<"\n";
xSrcPrev = xSrcCurr;
xRcvPrev = xRcvCurr;
//if (i < numSamples/2)
// g = - velRelative/SPEED_OF_SOUND; // "growth parameter"
//else
g = + velRelative/SPEED_OF_SOUND;
readPtr+=g; //update read pointer with variable delay
//wraparound checks:
if (readPtr < 0)
readPtr+=DELAY_SIZE;
if (readPtr>DELAY_SIZE-1)
readPtr-=DELAY_SIZE;
//write pointer simply increments...
writePtr++;
if (writePtr>DELAY_SIZE)
writePtr-=DELAY_SIZE;
//where the magic happens: the variable delay creates the frequency shift
delay.setDelay(1+readPtr);
//User Selectable switching of the "Mixing": gets rid of discontinuities
// when the read pointer crosses the write location by
// linearly mixing between half a delay line away.
// maximum amplitude is respected, but at the cost
// of additional artifacts. better mixing techniques could reduce these...
double mix = ( (int)(readPtr-writePtr) % (DELAY_SIZE) ) / (double) DELAY_SIZE ;
if (atoi(argv[7]) == 0)
mix = 1; // overrides the mixing calculation. i.e. no mixing.
//update output sample
float outsample = mix*delay.tick(inputLoop.tick());
//apply mixing (if any)
outsample += (1-mix)*delay.tapOut(DELAY_SIZE/2);
//put into frames
frames[i] = outsample;
//std::cout<<"tE = " << timeElapsed<<"\n";
}
//dump frames into outputFile file. (could have incrementally done it directly above...)
outputFile.tick(frames);
return 0;
}
int main()
{
StkFloat d; // sample data
unsigned long inputSize; // size of the input file in sample frames
double inputDuration; // duration of the input file loop
unsigned int channels = 1;
//set global sample rate
Stk::setSampleRate(44100.0);
int nFrames=100000;
FileLoop input;
FileWvOut output;
// open the input and output files, with error checking
try {
// load the input file
input.openFile("rawwaves/Noisy_Miner_chirp_stereo.wav", false);
// open a 16 bit, one channel WAV formatted output
output.openFile("many_chirps.wav", 1, FileWrite::FILE_WAV, Stk::STK_SINT16);
}
catch (StkError &) {
exit(1);
}
// Set input read rate based on the default STK sample rate.
double rate;
double inputRate;
inputRate = input.getFileRate();
rate = inputRate / Stk::sampleRate();
input.setRate( rate );
// Find out the length of the input file
inputSize=input.getSize();
inputDuration = inputSize/inputRate;
std::cout << "input file is " << inputSize << " samples long with a duration of " << inputDuration << " sec." << std::endl;
// Find out how many channels we have.
channels = input.channelsOut();
std::cout << "input file has " << channels << " channels at "<< inputRate << "Hz sample rate " << std::endl;
std::cout << " we want to run a rate of " << Stk::sampleRate() << " and will sample the file by a factor of " << rate << std::endl;
// we want the chirps to be at the same frequency as they were recorded
// so we adjust the looping rate of recorded loop
// if you make the frequency negative, it reverses the loop!
input.setFrequency(1.0/inputDuration);
// run the oscillator for 40000 samples, writing to the output file
for (int i=0; i<nFrames; i++){
try {
d=input.tick(); // read 1 sample from the input file
output.tick(d); // write 1 sample to the output file
}
catch (StkError &) {
exit(1);
}
// print the first 100 samples to the std output for debugging
if (i<10){
std::cout << d << std::endl;
}
}
return 0;
}