Scripts and configuration files for performing speech recognition on OGI's ISOLET corpus with the Quicknet multi-layer perceptron (MLP) package for acoustic modeling and the noway hidden Markov model (HMM) decoder can be downloaded here. A tutorial is included which is meant to make it easy for people new to Quicknet to use the scripts. The scripts support recognition on both the original ISOLET corpus and a noisy version which has background noise added. Support for multi-stream speech recognition using a different feature vector for each stream is built in.
In November 2008 we added new duration modeling options that can be used to improve speech recognition accuracy (see the README_DURATION file that we have included). In July 2008 we updated the README to make it clearer how to perform statistical significance testing, and added a binomial probability calculation tool to the testbed for significance testing. In June 2008 we added an important clarification to the README about how to do multi-threaded MLP training properly, without choosing a mlp3_bunch_size value that is too high. In May 2008 we added the mergePosteriors tool, which is used to merge streams in a multi-stream system. The old tool we used to do this was available only as a Linux binary, but the new mergePosteriors tool is a regular script like the other scripts. We updated the HCopy MFCC configuration file in March 2008 to improve the performance of MFCC features. See the file README_MFCC.
For more information about ISOLET, along with scripts and noise files that can be used to duplicate our noisy version of ISOLET see here. You can find scripts and configuration files for ISOLET speech recognition using the HTK toolkit instead of Quicknet at the same location. The Quicknet tools and noway can be downloaded here.