Audio Stream Analysis and Event Extraction Model

From Emcap

Contents

Ricaudio

by Ricard Marxer <email@ricardmarxer.com>

Music Technology Group, Universitat Pompeu Fabra

2008/2/4

Copyright (c) 2006-2008 All Rights Reserved.

Download Emerge 1.0 (GNU/Linux)

Download Emerge 1.0 (Windows)

What Is The Module?

It is the implementation of several onset detectors methods in order to split an audio stream into several audio events. The resulting segments are then optionally analyzed by aubio in order to extract a feature vector.

What Are The Inputs/Outputs?

This module takes as input a filename of a WAV audio file. It outputs a stream of feature events containing the timestamp (using onset detection models) and the feature vector (using sound analysis models). In the future the onset detection model used as well as the feature vector extracted will be controllable using parameters.

What Are The Parameters?

The parameters possible are the onset detection method and gain, as well as the features that the auditory model should extract.

Usage: emem-ricaudio [options]
Options:
 -h, --help            show this help message and exit
 -m ONSET_DETECTION_METHOD, --odf-method=ONSET_DETECTION_METHOD
                       onset detection function method: 'mkl', 'kl',
                       'spectral_flux', 'spectral_difference',      
                       'phase_deviation', 'phase_deviation_weighted',
                       'complex_domain', 'complex_domain_hypot' or 'hfc'
                       [default: complex_domain_hypot]                  
 -g GAIN, --gain=GAIN  the value for the threshold gain [default: 0.9]  
 -f FEATURES, --features=FEATURES                                       
                       set of features to extract: 'cochlea', 'mfcc', 'pitch'
                       or 'duration' [default: cochlea]

What Are The System Requirements?

GNU/Linux

The package can be installed in several different ways, depending on your system. The easiest way is to install the setuptools package. On Debian based systems:

$ sudo apt-get install python-setuptools

and then just write the following:

$ sudo easy_install http://www.ricardmarxer.com/emcap/Emerge-1.0-py2.5.egg

If this method does not work correctly, one must install manually all the dependencies, and build the package himself. For the software to fully work you must have the following dependencies installed:

Then one must download the package here and run:

$ python setup.py build
$ sudo python setup.py install

Windows

A binary package has been made for ease of installation and use. To run, one must unzip the package somewhere and run the following executable from the command line:

emem-ricaudio.exe [OPTIONS] INPUT_FILENAME OUTPUT_FILENAME

How To Cite

@incollection{marx07,
 author = {Ricard Marxer and Piotr Holonowicz and Hendrik Purwins},
 title = {Dynamical Hierarchical Self-Organization of Harmonic, Motivic, and
 Pitch Categories},
 address = {Vancouver, Canada},
 booktitle = {Music, Brain and Cognition. Part 2: Models of Sound and
 Cognition, held at NIPS)},
 note = {(to appear)},
 year = {2007}
}

How To Use

$ emem-ricaudio [OPTION] INPUT_FILENAME OUTPUT_FILENAME

Input File Format

The input file must be a WAV or FLAC audio file.

Output File Format

One feature event per line of the form:

ONSET-TIME,ONSET-TIME-VAR;FEAT1-VAL,FEAT1-VAR FEAT2-VAL,FEAT2-VAR ... FEATn-VAL,FEATn-VAR

Example Of Use

$ emem-ricaudio test.wav test.fev