Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!mcgill-vision!bloom-beacon!snorkelwacker!tut.cis.ohio-state.edu!cs.utexas.edu!uunet!pdn!rnms1!dave
From: dave@rnms1.paradyne.com (Dave Cameron (Consultant))
Newsgroups: comp.ai
Subject: Re: transcripts of conversations
Message-ID: <7342@pdn.paradyne.com>
Date: 6 Feb 90 16:34:16 GMT
References: <41296@lanl.gov> <814@jethro.Corp.Sun.COM>
Sender: usenet@pdn.paradyne.com
Reply-To: dave@rnms1.paradyne.com (Dave Cameron )
Organization: AT&T Paradyne, Largo, Florida
Lines: 8

In article <814@jethro.Corp.Sun.COM> acm@grendal.EBay.Sun.COM (Andrew MacRae) writes:
>If someone is looking for transcripts of face to face conversations,
>you might consider using play scripts.  Granted, depending on the playwrite
>the conversations may be less than realistic, but many plays do mimic
>normal conversation fairly well.

This is far from correct. If you are looking for strings of phonetic groups
it may do, but for word choice, content analysis, and pattern expectation