Gist - An Audio Analysis Library
Gist is a C++ based audio analysis library
Author
Gist is written and maintained by Adam Stark.
Usage
Firstly, import the Gist header file:
#include "Gist.h"
Instantiation
Gist is a template class, so instantiate it with floating point precision:
int frameSize = 512;
int sampleRate = 44100;
Gist<float> gist (frameSize, sampleRate);
Or with double precision:
Gist<double> gist (frameSize, sampleRate);
We proceed with the documentation as if we were using floating point precision.
Process Audio Frames
Once you have an audio frame, pass it to the Gist object. You can do this either as a STL vector:
std::vector<float> audioFrame;
// !
// fill audio frame with samples here
// !
gist.processAudioFrame (audioFrame);
Or, as an array:
float audioFrame[512];
// !
// fill audio frame with samples here
// !
gist.processAudioFrame (audioFrame, 512);
Now we can retrieve some audio features.
Core Time Domain Features
// Root Mean Square (RMS)
float rms = gist.rootMeanSquare();
// Peak Energy
float peakEnergy = gist.peakEnergy();
// Zero Crossing rate
float zcr = gist.zeroCrossingRate();
Core Frequency Domain Features
// Spectral Centroid
float specCent = gist.spectralCentroid();
// Spectral Crest
float specCrest = gist.spectralCrest();
// Spectral Flatness
float specFlat = gist.spectralFlatness();
// Spectral Rolloff
float specRolloff = gist.spectralRolloff();
// Spectral Kurtosis
float specKurtosis = gist.spectralKurtosis();
Onset Detection Functions
// Energy difference
float ed = gist.energyDifference();
// Spectral difference
float sd = gist.spectralDifference();
// Spectral difference (half-wave rectified)
float sd_hwr = gist.spectralDifferenceHWR();
// Complex Spectral Difference
float csd = gist.complexSpectralDifference();
// High Frequency Content
float hfc = gist.highFrequencyContent();
FFT Magnitude Spectrum
// FFT Magnitude Spectrum
const std::vector<float>& magSpec = gist.getMagnitudeSpectrum();
Pitch
// Pitch Estimation
float pitch = gist.pitch();
Mel-frequency Representations
// Mel-frequency Spectrum
const std::vector<float>& melSpec = gist.getMelFrequencySpectrum();
// MFCCs
const std::vector<float>& mfcc = gist.getMelFrequencyCepstralCoefficients();
Version History
=== 1.0.4 === (22nd January 2017)
- Small changes to the interface for MFCCs and Mel Spectrum
- Many internal improvements
- Added window functions
=== 1.0.3 === (17th June 2016)
- Added a Python module
- Added ability to set and get sampling frequency
- Bug fixes and implementation improvements
=== 1.0.2 === (24th April 2016)
- Added the option of using Apple Accelerate FFT
- Added two new features: Spectral Rolloff and Spectral Kurtosis
- Small usability and code style tweaks
=== 1.0.1 === (26th June 2014)
- Added the option of using Kiss FFT instead of FFTW
=== 1.0.0 === (22nd June 2014)
- The first version of Gist
Dependencies
The Gist library depends on one of the following FFT libraries:
You will need to install this yourself, link projects using -lfftw3 and use the flag -DUSE_FFTW
- Kiss FFT - included with project
This is included with the project. To use Kiss FFT, add the flag -DUSE_KISS_FFT
To use Accelerate FFT, add the flag -DUSE_ACCELERATE_FFT
License
Copyright (c) 2014 Adam Stark
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/.