comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado
Home » Public Forums » archive » Modelling the PSD using the EM algorithm
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend  		Switch to threaded view of this topic Create a new topic Submit Reply
Modelling the PSD using the EM algorithm [message #14042] Tue, 19 January 1999 00:00
Mark McGillion is currently offline  Mark McGillion
Messages: 14
Registered: December 1997
Junior Member
All,

This is a long shot but I thought I would give it a try any how.

I have a power spectral density function (PSD) computed from a short
time segment of speech. I am trying to reduce the dimensionality
(currently 1000 points) in order to train a Neural Network (ANN) to
recognise speech pattern sequences.

By treating the PSD as a 1 dimensional probability density function, my
approach is to model the PSD using a mixture of Gaussians and use the
parameters of the distributions to train the ANN. The standard method is
to use maximum likelihood an implementation of which is the EM algorithm
(Dempster et al).

Does anybody know if this has been done before or if there is any
documentation/IDL code out there regarding this?

Regards,

Mark

Report message to a moderator

  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: IDL FAQ Update
Next Topic: Re: XMANAGER with not top level base?

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 18:40:35 PDT 2025

Total time taken to generate the page: 0.00622 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.7.
Copyright ©2001-2013 FUDforum Bulletin Board Software