Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!pipex!uunet!majipoor.cygnus.com!crash
From: crash@cygnus.com (Jason Molenda)
Subject: What physical grounding do formants have?
Message-ID: <CJFCuw.124@cygnus.com>
Organization: Cygnus Support
Date: Mon, 10 Jan 1994 17:35:18 GMT
Lines: 18

howdy, I picked up Rabiner and Juang's _Fundamentals of Speech
Recognition_ Friday after reading the FAQ -- great book so far (I'm
only just now getting into chapter three).  My only gripe would be that
the discussion of time delay neural networks and hidden control neural
networks was, uh, rather brief (at the end of chapter two).  I thought
the exercise half-way through chapter two, where they show you the
spectrograms for the ten counting digits and ``oh'' and have you try to
figure them out by only looking at the spectrograms, was a great idea
(I failed miserably but it was an interesting exercise).


One thing about which I was not clear -- is the sound for a particular
formant tied to a particular part of the vocal tract, i.e. the
frequencies in the first formant (F1) are made between the glottis and
velum?

Sorry for the really basic question; before Friday I didn't know
anything about this stuff.  I now know marginally more than nothing.  :-)
