The Hidden Markov Model Toolkit (HTK) is a software toolkit for building Hidden Markov Model (HMM) speech recognition and classification systems. HTK is ideal for research in HMM training, modeling and speech recognition. HTK is built on an extensible, modular library design that simplifies creation of user developed tools. HTK includes many signal processing functions, HMM training and testing tools, as well as language modeling support and recognizer scoring. HTK can be integrated with ESPS and waves+. HTK requires IRIX OS 4.0.5F or greater.
Ken Nelson
Director, Marketing and Sales Entropic Research Laboratory, Inc. 600 Pennsylvania Ave. SE Suite 202 Washington, DC 20003 USA 202-547-1420 fax:202-546-6648