Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!yeshua.marcam.com!uunet!munnari.oz.au!cs.mu.OZ.AU!miscord
From: miscord@ee.mu.OZ.AU (Michael Scordilis)
Subject: Re: digital inverse filtering
Message-ID: <9426314.4353@mulga.cs.mu.OZ.AU>
Sender: news@cs.mu.OZ.AU
Organization: Department of Electrical & Electronic Engineering, University of Melbourne
References: <3571su$t7v@galileo.polito.it>
Date: Tue, 20 Sep 1994 04:05:54 GMT
Lines: 17

In article <3571su$t7v@galileo.polito.it> TELTESI22@POLCLU (Eric Lora Ronco) writes:
>In the article "An Algorithm for Automatic Formant Extraction Using Linear
>Prediction Spectra" from S. McCandless, lpc coefficients are computed using
>the digital inverse filtering technique. The reference is:
>___________________
>Eric Lora Ronco


This relates to how the system identification problem is set up.
It can be viewed as inverse filtering where the input is speech, the output
is the excitation sequence and the system is, A(z) the corresponding inverse
filter of the system.  The resulting solution is identical to any other solution
(all other factors being equal), in the sense that it will give same results.

M  Scordilis

Michael Scordilis	   	|internet    miscord@mullian.ee.mu.OZ.AU
