BTrack - A Real-Time Beat Tracker

** Version 1.0.4 **

by Adam Stark, Matthew Davies and Mark Plumbley.

About BTrack

BTrack is a causal beat tracking algorithm intended for real-time use. It is implemented in C++ with wrappers for Python and the Vamp plug-in framework.

Full details of the working of the algorithm can be found in:

• Musicians and Machines: Bridging the Semantic Gap in Live Performance, Chapter 3, A. M. Stark, PhD Thesis, Queen Mary, University of London, 2011.

• Real-Time Beat-Synchronous Analysis of Musical Audio, A. M. Stark, M. E. P. Davies and M. D. Plumbley. In Proceedings of the 12th International Conference on Digital Audio Effects (DAFx-09), Como, Italy, September 1-4, 2009.

BTrack is made available under the GNU General Public License, version 3. Please see the included LICENSE.txt for more details.

Versions

==== 1.0.4 ==== (18th June 2016)

• Added option of using Kiss FFT
• Implementation changes to improve efficiency

==== 1.0.3 ==== (10th January 2016)

• Fixed implementation error in complex spectral difference (thanks to @zbanks for pointing it out)
• Small change to python module build script

==== 1.0.2 ==== (25th November 2014)

• Added updated Max external project and associated files
• Fixed a couple of bugs

==== 1.0.1 ==== (21st November 2014)

• Moved to git & updated README
• No implementation changes

==== 1.0.0 ==== (8th July 2014)

• Many updates to stability and improvements to implementation

==== 0.9.0 ==== (circa 2008/2009)

• This is the original version of the BTrack algorithm

Usage - C++

STEP 1

Include the BTrack header file as follows:

#include "BTrack.h"

STEP 2

Instantiate the algorithm by one of the following:

// to use the default 512 hop size and 1024 frame size
BTrack b; 

or:

// to specify a hop size (e.g. 512) and have a frame size of 2 x the hop size
BTrack b(512); 

or:

// to specify both the hop size and frame size
BTrack b(512,1024);

STEP 3.1 - Audio Input

In the processing loop, fill a double precision array with one frame of audio samples (as determined in step 2):

double *frame; 

// !
// do something here to fill the frame with audio samples
// !

and then call:

b.processAudioFrame(frame);

and to check for beats, simply call:

if (b.beatDueInCurrentFrame())
{
	// do something on the beat
}

STEP 3.2 - Onset Detection Function Input

The algorithm can process onset detection function samples. Given a double precision onset detection function sample called 'newSample', at each step, call:

b.processOnsetDetectionFunctionSample(newSample);

and then check for beats with:

if (b.beatDueInCurrentFrame())
{
	// do something on the beat
}

Requirements

To compile BTrack, you will require the following libraries:

• libsamplerate

And either:

• FFTW (add the flag -DUSE_FFTW)

or:

• Kiss FFT (included with project, use the flag -DUSE_KISS_FFT)

License

Copyright (c) 2014 Queen Mary University of London

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 3 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, see http://www.gnu.org/licenses/.