Xref: utzoo comp.ai:9177 comp.sources.wanted:16486 sci.math.stat:2278
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!uunet!munnari.oz.au!brolga!aas
From: aas@brolga.cc.uq.oz.au (Alex Sergejew)
Newsgroups: comp.ai,comp.sources.wanted,sci.math.stat
Subject: Looking for Markov modelling code (non-external models from time series)
Message-ID: <1991May2.113425.802@brolga.cc.uq.oz.au>
Date: 2 May 91 11:34:25 GMT
Organization: Prentice Computer Centre, The University of Queensland, Australia.
Lines: 18

The header pretty much says it all.  I'm aware that there is an extensive
literature on efficiently estimating state transitions and unknown parameters
of hidden and semi markov signal models from time series data, primarily used
in the speech processing context, but have been making heavy weather of
translating the basic references I've found into useful working code.
Looking in the usual places (netlib, Numerical recipes, archie, etc) has drawn
a blank, and all the speech processing types I've been able to ask seem not
to have code or algorithms they deem fit for sharing.

Could anyone please point me in the right direction, whether it be algorithmic
descriptions or (preferably) PD source code.  If there is enough interest
I will gladly post a summary.

Alex.

 _--_|\    Alex A Sergejew               Internet:  aas@cc.uq.OZ.AU
/      X   University of Queensland      Voice:     +61-7-271-8298
\_.--._/   Brisbane, Qld, Australia      Fax:       +61-7-271-8567