We describe the maximum-likelihood parameter estimation problem and how the Expectation- Maximization (EM) algorithm can be used for its solution. We first describe the abstract form of the EM algorithm as it is often given in the literature. We the
dhmm = HMM with discrete output mhmm = HMM with mixture of Gaussians output; Use mhmm with M=1 components to simulate an HMM with a single Gaussian output.
(Adapted from Hidden Markov Toolbox Version 3.3 01-Apr-99 and Coupled Hidden Markov Toolbox Version 1.1 01-Feb-01 Copyright (c) by Iead Rezek, Oxford University) The software uses some NETLAB routines so you'll need to have NETLAB on your search pat
demgausshmm.m uses Gaussian observation model demgausshmm_traj.m uses Gaussian observation model on a trajectory demgausshmm_back.m Gaussian observation model using backwards compatibility functions dempoissonhmm.m uses Poisson observation model on
C11 objects of types that have virtual functions or virtual base classes contain volatile (‘memory’) pointers. We call such pointers ‘hidden pointers’ because they were not specified by the user. If such C11 objects are made persistent, then these p
HTK, Hidden Markov Model Toolkit. HTK source code, ver 3.4.1 (tar+gzip archive), for Linux/Unix. HTK was originally developed at the Cambridge University Engineering Department (CUED).