Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!doc.ic.ac.uk!agate!howland.reston.ans.net!europa.eng.gtefsd.com!paladin.american.edu!zombie.ncsc.mil!afterlife!jpcampb
From: jpcampb@afterlife.ncsc.mil (Joe Campbell)
Subject: Re: HMM query
Message-ID: <1994Jan18.070121.24500@afterlife.ncsc.mil>
Organization: The Great Beyond
References: <9401121152.AA06747@dcs.shef.ac.uk>
Date: Tue, 18 Jan 1994 07:01:21 GMT
Lines: 19

In article <9401121152.AA06747@dcs.shef.ac.uk>,
Dorota Iskra <D.Iskra@dcs.shef.ac.uk> wrote:
>I need help. I want to train HMM'S USING FORMANT INFORMATION.
>Somebody must have done it before, but I have not been able
>to find any references.  Does anybody know of anything published
>on this subject?

I'm sorry I don't have answers to your questions off the top
of my head, but I have some questions...  Do you know of a
reliable and robust formant estimation algorithm?  How do
you propose handling the varying number of formants in your
HMM?

Joe
-- 
.............................................................................
; Dr. Campbell  N3JBC  jpcampb@afterlife.ncsc.mil  74040.305@compuserve.com ;
; I speak for myself!     Happiness = Reality - Expectations, Click & Clack ;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
