LINGUIST List 3.52

Tue 21 Jan 1992

Disc: IPA, Parsers, Voice Transcription, Fonts

Editor for this issue: <>

Directory

1. "E_Dean.Detrich", 3.36 Queries: Computer Transcrip
2. Atro Voutilainen, 3.25 Queries: Tagger, ASL, Dutch, Mabang, Eskimo
3. Michel Eytan LILoL, Re: 3.36 Queries: Computer Transcription, Wordstar, Verlan
4. meSUZUKA.U-STRASBG.FR(, IPA fonts for WP (again, alas)

With regard to Vicki Fromkin's query about IPA for Wordstar I can say this.
There is a company in Houston, TX called M.A.P. Systems which has prepared a
program called "Lines, Boxes, Etc - Phonetic Edition" for Wordperfect 5.1
whichI am currently using and which is very nice. In addition they have a
version called "Type, See and Print - Phonetic Edition" which is used with
programs other than WP. I do not know how convenient the latter is to use,
but if it is of the same quality as the WP version, it is quite good.

Contact: M.A.P. Systems
 18100 Upper Bay Road #100
 Houston, Texas 77058

 (800) 527 2851 or (713) 333 9640 or fax (713) 333 9579

E. Dean Detrich
Michigan State University

P.S. They have been most helpful on the telephone solving installation
problems.

> 1)
> Date: Thu, 9 Jan 92 10:31:07 EST
> From: rohinics.Buffalo.EDU (Rohini Srihari)
> Subject: Part-of-Speech Tagger
>
> I wanted to know where I could obtain a part-of-speech
> tagger. In particular, I am looking for the one
> described by Church in a 1988 paper. I would appreciate
> information on other available POS taggers as well.
>
> Thanks,
>
> Rohini K. Srihari
> Center for Document Analysis and Recognition
> SUNY at Buffalo
>

Hopefully by mid-1992, ENGCG, a tagging and parsing system for
English, documented in

Karlsson, Voutilainen, Anttila & Heikkila (1991),
	Constraint Grammar: a Language-Independent System for
	Parsing Unrestricted Text, with an Application to
	English. Natural Language Text Retrieval: Workshop Notes
	from the Ninth National Conference on Artificial
	Intelligence (AAAI-91). Anaheim, California.

Karlsson, Voutilainen, Heikkila & Anttila (book forthcoming in 1992),
	Constraint Grammar: a Language-Independent System for
	Parsing Unrestricted Text.

will become publicly available. This system performs not only plain
parts-of-speech disambiguation but also shallow dependency-oriented
functional syntactic analysis. At the level of disambiguation, at least
99.7% of all words get an appropriate reading. On the other hand, some
3-6% of the ambiguities remain (partly) unresolved.

	For more details, contact any of the authors

Fred Karlsson fkarlssoLing.Helsinki.FI
Atro Voutilainen avoutilaLing.Helsinki.FI
Juha Heikkila jheikkilLing.Helsinki.FI
Arto Anttila aanttilaLing.Helsinki.FI

-- Atro Voutilainen

Message 3: Re: 3.36 Queries: Computer Transcription, Wordstar, Verlan

>Date: Tue, 14 Jan 1992 10:35 EST
>From: Barbara Ruth Campbell <CAMPBELLzodiac.rutgers.edu>
>Subject: IPA and Computer Transcription

>I don't know if this question will sound too unrealistic or not but after
>seeing segments on computer voice recognition systems I was wondering
>whether anyone has worked or plans on working on a computer system that
>would have people speak into a recorder and then the computer would
>encode the phonetic "bits" into IPA
...
>Any comments? Is there such a system? If not, could one be devised?
>Does anyone think that such a system might help in accent modification or
>learning a target language or teaching an actor to emulate the accent of
>a target area?

if you can work out the system you describe, it would be very useful. As you
 state it, it would be quite complex because of the vocal side of things. I am
 however thinking of another application: writing/transferring files by e-mail
 in non-standard -- ie non English -- alphabets, ranging from Latin but
 accentuated (French, German, in fact all other european) ones to non Latin
 alphabets (eg, Hebrew, Greek, Russian) or even to ideogrammatic ones (Chinese,
 Japanese, etc). I have heartd somewhere that there is a proposal to set the
 ASCII standard at 32 bits, but till that comes along (if it does, since it will
 be a heavy load on the net) transcription might be useful -- although it has
 its own set of problems, specially the regional accents, dialects, nay the many
 _different_ languages using a common set of ideograms (cf Chinese again) or a
 common alphabet (eg Hebrew and Yiddish).

==michel eytandpt-info.u-strasbg.fr

there has been some trouble involved downloading the file drivers.exe. My only
 guess is that it happened to netters not working under unix.

I have worked out a solution: I can send the files by e-mail, from a PC to
 another PC on the explicit condition that both of us have POPmail
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
this (a freeware mailer) can be obtained by anonymous ftp from
 boombox.micro.umn.edu (or 128.101.95.95, but this can change, whereas the name
 does not change). Write to ask for transfer via POPmail.

Finally, as a last-ditch stand: if you send me a self-addressed envelope
 (padded, preferrably) with enough 'international postal coupon'-s and a virgin
 720K-diskette I shall copy the files and send it back to you.

==eytandpt-info.u-strasbg.fr