Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!yeshua.marcam.com!charnel.ecst.csuchico.edu!csusac!csus.edu!netcom.com!olea
From: olea@netcom.com (Michael Olea)
Subject: Re: wavelets and NN
Message-ID: <oleaCw8M6t.LHC@netcom.com>
Keywords: wavelets, associative memory
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <35987j$r4n@kodak.rdcs.Kodak.COM>
Date: Fri, 16 Sep 1994 19:30:29 GMT
Lines: 36

martin@belteshazzar.Kodak.COM (Craig Martin) writes:


>-- 
>I am currently working on a thesis project in speech recognition that
>uses wavelet tranforms to generate the feature vector and then training
>some of the vectors into associative memory (NN).  I have recorded
>15 people saying the vowels from the Peterson/Barney group to use as
>my sound data base.  Some of these are trained into the associative
>memory and some are presented to the network for recognition.
> 
>My problem is that I cannot seem to generate good, distinct, feature
>vectors using the wavelet (DAUB4 as shown in numerical recipes).  My
>approach has been to parse the vowel from the beginning of each sound
>into a 1024 byte buffer, normalize the data, feed it into the wavelet
>algorithm, and then add up the coefficients in each section of the
>resulting wavelet transform (first 2, second 2, next 4, next 8...etc).
>There is some similarity in the summed coefficients for a given sound
>spoken by different people, but not enough to give me distinct feature
>vectors for a given sound.
> 
>I have tried the higher order wavelet transforms from the nr book with
>no improvement.
> 
>Is there anyone out there that can tell me what I am doing wrong?????
>It is probably a lack of understanding on my part as to how to interpret
>the wavelet transform.
> 
>Thanks in advance.......
>-- 
>Craig W. Martin                            martin@belteshazzar.Kodak.COM

	Try: Eva Wesfreid and Mladen Victor Wickerhauser, "Adapted local
trigonometric transform and speech processing" in  IEEE Transactions on
Signal Processing, 41(12):3596-3600, December 1993.

