Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!pipex!sunsite.doc.ic.ac.uk!lyra.csx.cam.ac.uk!warwick!uknet!bcc.ac.uk!pc13.phon.ucl.ac.uk!andrew
From: andrew@PROBLEM_WITH_INEWS_DOMAIN_FILE ()
Subject: Re: categorical perception experiment
Sender: news@ucl.ac.uk (Usenet News System)
Message-ID: <1994Dec15.102402.31373@ucl.ac.uk>
Date: Thu, 15 Dec 1994 10:24:02 GMT
References: <3cl007$hf@lyra.csx.cam.ac.uk>
Organization: I need to put my ORGANIZATION here.
X-Newsreader: TIN [version 1.2 PL2]
Lines: 79

res23 (res23@cam.ac.uk) wrote:
: Hello,
:   I have been directed to this newsgroup and I hope you will be able to
: help me with a research question I have.  I am hoping to use a typical
: categorical perception linguistic task in a project I am doing.  This
: would involve creating artificial phonemes which vary along voice onset
: time, formant frequency, etc. continua, and I am looking for software
: to use to generate the stimuli.  I have looked in the FAQ, and I saw a
: couple programs listed that look like they might be useful - ie. sensyn
: speech synthesizer and macintalk were two that looked possible.  The
: ones that just generate natural speech won't do, since I need to vary
: the precise characteristics of the sounds, but these programs sounded
: like they might have this capability.  Has anyone used these programs?
: Any comments? Or ideas about what might be the best way to generate
: this task? I'll be using a Mac, probably a Quadra 610, possibly with an
: AudioMedia sound card, or possibly a powerMac.
:   Thanks for your help.
: Robyn

Unfortunately this solution is not good for the Macintosh, but if you
have access to a PC or a Sun Sparc Station....


KPE - A Klatt Parameter Editor
==============================

This program provides a simple graphical interface for the implementation
of the Klatt 1980 synthesiser provided by Jon Iles
(j.p.iles@cs.bham.ac.uk) and Nick Ing-Simmons (nicki@lobby.ti.com).

It was inspired by IGE, a piece of code written by Rob Fletcher
(rpf1@unix.york.ac.uk).

The comprises an X-Windows interface written using the SUIT 
toolkit and version 3.03 of the synthesiser code.

The interface allows users to display and edit Klatt parameters using
a graphical display which includes the time-amplitude waveform of the
resultant synthetic speech.  The code was written to help in the
production of copy syntheses of natural speech tokens, and so also
includes the facility to display the original speech waveform together
with its synthetic copy.  In addition to being able to play back both
original and synthetic stimuli and match them 'by ear', it is possible
to compare short term spectra of regions of both stimuli.

An example of a natural speech token, and a rough synthesised copy are
included.

The code has been written to use the Speech Filing System (SFS)
available from UCL; this can be obtained via anonymous FTP.  This was
used to calculate the initial parameter value estimates.

I should stress that the code is supplied 'as is' and no promises are
made about its reliability.  It would be nice to get feedback on it,
and to develop it further.

The code has been developed on a Sun Sparc10, but should work on a
Sparc2 as well, although the audio output won't be as good.  A Linux
version is also available.

Where to get the code:

via anonymous ftp from:

pc13.phon.ucl.ac.uk:/pub/kpe/kpe80.sun4.tar	KPE for sun4 systems
pc13.phon.ucl.ac.uk:/pub/kpe/kpe80.linux.tar	KPE for linux systems

pitch.phon.ucl.ac.uk:/pub/sfs:			The SFS distribution.

kpe@pc13.phon.ucl.ac.uk
http://www.phon.ucl.ac.uk/

Apologies for minimal documentation. Feedback welcome.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.


