Newsgroups: comp.speech
Path: pavo.csi.cam.ac.uk!doc.ic.ac.uk!uknet!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!sdd.hp.com!cs.utexas.edu!csc.ti.com!tilde.csc.ti.com!ra.csc.ti.com!pluto!a904209
From: nicki@lobby.ti.com (Nick Ing-Simmons)
Subject: pre-release of "public domain" text-to-speech code
Message-ID: <CFzCJ4.416@csc.ti.com>
Sender: usenet@csc.ti.com
Nntp-Posting-Host: pluto.tiuk.ti.com
Reply-To: nicki@lobby.ti.com
Organization: Texas Instruments Ltd.
Date: Thu, 4 Nov 1993 18:25:52 GMT
Lines: 27

I have been trying to pull together the various parts of a 
text to speech system from various public domain sources.

A pre-release which works on Sun SPARC stations (which is all
I have access to can be found on the comp.speech archive site i.e.

svr-ftp.eng.cam.ac.uk

In /pub/comp.speech/sources/rsynth-0.9.tar.Z

This several steps beyond the "speak" program which concatenates .au files,
but is not yet "finished".  The bottom level is based on the "klatt-0.02"
code posted here some time back (heavily modified).

However I have not got much (any?) time to work on this at present,
so I am posting it in the hope that it may help some people as-is.

I would appreciate feedback as to how it compares 
(strengths/weaknesses) with other systems, and any "expert" advice on
how to improve it, as and when I get time. 

Nick Ing-Simmons nicki@lobby.ti.com

(It is a private project and only connection with TI is that I work
 for them - but not on this.)


