/*
* _______ _____ _____ _____
* |__ __| | __ \ / ____| __ \
* | | __ _ _ __ ___ ___ ___| | | | (___ | |__) |
* | |/ _` | '__/ __|/ _ \/ __| | | |\___ \| ___/
* | | (_| | | \__ \ (_) \__ \ |__| |____) | |
* |_|\__,_|_| |___/\___/|___/_____/|_____/|_|
*
* -------------------------------------------------------------
*
* TarsosDSP is developed by Joren Six at IPEM, University Ghent
*
* -------------------------------------------------------------
*
* Info: http://0110.be/tag/TarsosDSP
* Github: https://github.com/JorenSix/TarsosDSP
* Releases: http://0110.be/releases/TarsosDSP/
*
* TarsosDSP includes modified source code by various authors,
* for credits and info, see README.
*
*/
package be.tarsos.dsp.resample;
import be.tarsos.dsp.AudioEvent;
import be.tarsos.dsp.AudioProcessor;
/**
* Sample rate transposer. Changes sample rate by using interpolation
*
* Together with the time stretcher this can be used for pitch shifting.
* @author Joren Six
*/
public class RateTransposer implements AudioProcessor {
private double factor;
private Resampler r;
/**
* Create a new sample rate transposer. The factor determines the new sample
* rate. E.g. 0.5 is half the sample rate, 1.0 does not change a thing and
* 2.0 doubles the samplerate. If the samples are played at the original
* speed the pitch doubles (0.5), does not change (1.0) or halves (0.5)
* respectively. Playback length follows the same rules, obviously.
*
* @param factor
* Determines the new sample rate. E.g. 0.5 is half the sample
* rate, 1.0 does not change a thing and 2.0 doubles the sample
* rate. If the samples are played at the original speed the
* pitch doubles (0.5), does not change (1.0) or halves (0.5)
* respectively. Playback length follows the same rules,
* obviously.
*/
public RateTransposer(double factor){
this.factor = factor;
r= new Resampler(false,0.1,4.0);
}
public void setFactor(double tempo){
this.factor = tempo;
}
@Override
public boolean process(AudioEvent audioEvent) {
float[] src = audioEvent.getFloatBuffer();
//Creation of float array in loop could be prevented if src.length is known beforehand...
//Possible optimization is to instantiate it outside the loop and get a pointer to the
//array here, in the process method method.
float[] out = new float[(int) (src.length * factor)];
r.process(factor, src, 0, src.length, false, out, 0, out.length);
//The size of the output buffer changes (according to factor).
audioEvent.setFloatBuffer(out);
//Update overlap offset to match new buffer size
audioEvent.setOverlap((int) (audioEvent.getOverlap() * factor));
return true;
}
@Override
public void processingFinished() {
}
}