[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Building Psychoacoustic Analysis Software

Dear Auditory List -

We are currently building software that analyses samples of sound or music using psychoacoustic and acoustic methods. The program is a template which centralises control of various routines for batch analysis, followed by appropriate statistical and graphical analysis. By using Matlab as the central environment we aim to exploit its graphing, statistical, and `soft computing' tools. The software will be designed with research in mind.

We are looking for analysis routines that provide descriptive parameters about aspects of spoken, sung or played sound primarily. Some of the types of descriptive parameters we are interested in include:

Psychoacoustic Models (Loudness, Sharpness, Roughness, Fluctuation Strength etc.)
Voice/Singing Analysis
Musical Phrase Analysis
Musical Texture Analysis
Beat Tracking and Tempo Analysis
Vibrato Analysis
Pitch and Pitch Strength
Binaural Spatial Measures

We are hoping that researchers on this list might be interested in contributing routines they may have implemented (not necessarily only Matlab code) for incorporation in this program. The hope is that once the software is relatively bug free it will be freely released, and that it may be useful for other researchers. Contributers would understandably expect some credit for supplying their software which is likely to be achieved through a citation system. References will be displayed when each parameter is calculated, so that users are reminded who developed the particular routine. The user manual for this software may also include (copyright permitting) any papers (or perhaps links to papers) associated with any contributed code.


Sam Ferguson
Densil Cabrera
Jens Brøsbol

Acoustics Research Laboratory
Faculty of Architecture
The University of Sydney