Hidden Markov Model (HMM) Toolbox written by Kevin Murphy (1998).
See http://www.ai.mit.edu/~murphyk/Software/hmm.html for details.
This version was last updated on 22 January 2003.

Installation
------------

1. Install netlab from http://www.ncrg.aston.ac.uk/netlab
2. Install KPMtools and KPMstats from
  http://www.ai.mit.edu/~murphyk/Software/index.html
3. Assuming you installed all these files in your matlab directory, In Matlab type

addpath matlab/netlab
addpath matlab/KPMtools
addpath matlab/KPMstats
addpath matlab/HMM
addpath matlab/HMM/Demos

Demos
-----

The following demonstration files are included.
They do not produce any interesting output; they merely check
the code runs without crashing. Please read the source code for
details.

- learn_dhmm_demo illustrates how to learn an HMM with discrete observations
- learn_mhmm_demo illustrates how to learn an HMM with a mixture of Gaussians observations
- fixed_lag_demo  illustrates how to do fixed lag smoothing
- online_em_demo  illustrates how to combine fixed-lag smoothing and EM


Use mhmm with M=1 components to simulate an HMM with a single Gaussian output.

References
-----------

See "A tutorial on Hidden Markov Models and selected applications in speech recognition",
 L. Rabiner, 1989, Proc. IEEE 77(2):257--286.
