Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!newsxfer.itd.umich.edu!gatech!howland.reston.ans.net!math.ohio-state.edu!newsfeed.acns.nwu.edu!ftpbox!mothost!lmpsbbs!news
From: corrigan@ukraine.corp.mot.com (Jerry Corrigan)
Subject: ASCII encoding of the IPA
Reply-To: corrigan@corp.mot.com
Organization: Motorola Chicago Corporate Research Labs
Date: Mon, 8 May 1995 15:45:32 GMT
Message-ID: <1995May8.154532.15687@lmpsbbs.comm.mot.com>
Sender: news@lmpsbbs.comm.mot.com (LMPSBBS News Account)
Nntp-Posting-Host: 182.1.83.60
Lines: 23


First, I'd like to thank everyone who responded to my request for information
on ASCII encodings of the IPA.  Your suggestions were very helpful.

There appear to be at least four systems that encode the complete IPA, with
diacritics.  These include the system by John Wells of University College London
(posted to this group a few days ago), a system by Evan Kirschenbaum of Hewlett-Packard
Laboratories (posted occasionally to this group and to sci.lang), one by David Branner
of the University of Washington, and Worldbet by James Hieronymus of AT&T Bell Laboratories.
This last is being used by the Center for Spoken Language Understanding at the Oregon
Graduate Institute for the speech data they distribute.  (I obtained the description
of the Worldbet at ftp://speech.cse.ogi.edu/pub/docs/worldbet.ps.Z)

There is also a numeric system recommended by the IPA for computer coding, and
various systems such as the arpabet (English) and SAM-PA (European languages) which
are useful for only a single or a particular set of languages.

Once again, thanks to everyone who responded.

Jerry Corrigan
corrigan@ukraine.corp.mot.com


