Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!warwick!pipex!uknet!mcsun!Germany.EU.net!gmd.de!newsserver.jvnc.net!howland.reston.ans.net!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!pasteur!wooters
From: wooters@ICSI.Berkeley.EDU (Chuck Wooters)
Subject: Word models available
Message-ID: <1993Apr20.212554.12677@pasteur.Berkeley.EDU>
Summary: Most likely pronunciations from TIMIT
Sender: nntp@pasteur.Berkeley.EDU (NNTP Poster)
Nntp-Posting-Host: icsib32.icsi.berkeley.edu
Organization: ICSI (International Computer Science Institute)
Date: Tue, 20 Apr 1993 21:25:54 GMT
Lines: 36

Hello-

I have extracted the most likely pronunciation for each of about 6100
words in the TIMIT database. If you would like to get these
pronunciations, they are available via anonymous ftp from
ftp.icsi.berkeley.edu in the directory pub/speech. The file is called
"TIMIT.mostlikely.Z". Here is a small piece of that file:

10 a
ax 
11 abbreviate
ix bcl b r iy v iy ey dx 
12 abdomen
ae bcl d ix m ix n 
13 abides
q ax bcl b ay dcl z 

Each entry if made up of two lines. The first line has the word number
followed by the spelling of the word. The second line contains the
transcription of the word using the set of 61 TIMIT phones.

A little clarification:
The transcriptions are taken directly from the hand-labeled
TIMIT database. As you might imagine, many words occured several
times in the database. Not every instance of a word was transcribed
in exactly the same way. So what I did was to take the transcription
that occured most often. That is what is in this file.

If you have any questions or comments, please let me know. I hope
this is helpful to someone.

-Chuck Wooters

-- 
Chuck Wooters
wooters@icsi.berkeley.edu
