ICSI LogoQuickNet

QuickNet is a suite of software that facilitates the use of multi-layer perceptrons (MLPs) in statistical pattern recognition systems. It is primarily designed for use in speech processing but may be useful in other areas.

The main pieces of the QuickNet package are:

QuickNet was developed in the Speech Group at the International Computer Science Institute by David Johnson. Major contributors include Dan Ellis, Chris Oei, Chuck Wooters and Philipp Faerber. Nelson Morgan provided the inspiration and algorithmic guidance and Krste Asanovic assisted with optimization. And last but not least, QuickNet wouldn't be what it is today without BoB!

QuickNet is open source software - license details are laid out in the COPYING file in the distribution. Some aspects of QuickNet may be protected by patents.

The latest version of QuickNet is availabe for download by FTP. For higher performance it is possible to compile QuickNet to use the ATLAS BLAS libraries. To run the testsuite code, which may or may not compile cleanly at any give time, the unmaintained rtst library is needed, along with some example data files. The feacat and pfile_utils and feacalc packages are also useful for pre and post-processing of data files.

Back to Tools page.