NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/speech/276

Internet Message Format | 1992-11-19 | 1.7 KB

Xref: sparky comp.speech:276 comp.dsp:2600 sci.lang:8100
Newsgroups: comp.speech,comp.dsp,sci.lang
Path: sparky!uunet!wupost!darwin.sura.net!uvaarpa!liberty!urbacht
From: [e-mail] (Tom Urbach)
Subject: Speech lab questions
Message-ID: <[e-mail]>
Date: Thu, 19 Nov 1992 20:42:29 GMT
Organization: Washington & Lee University
X-Newsreader: Tin 1.1 PL5
Lines: 27

I'm looking into setting up a microcomputer or workstation based
speech lab here at my uni and would like to get suggestions and
recommendations. References to relevant FAQ archives would also be a
help.

The primary use for the lab would be instruction and psych research in
human speech production/perception. In addition to analysing and
visualizing speech sounds in the usual ways, e.g., spectrograms, I'd
like to be able to prepare speech sounds as stimuli for psych
experiments--categorical perception, phoneme restoration, cross-modal
priming, etc. This means being able to mix two signals w/ millisecond
accuracy, e.g., superimposing a timing signal on a voice track or
replacing a phoneme w/ a click. It would also be nice to be able to
transform the signal in various ways, e.g., to devoice a phoneme here,
to pull out F3 there, to vary pitch while keeping speech rate
constant, and to vary speech rate while keeping pitch constant.

Any tips on hardware and software combinations would be greatly
appreciated, particularly from the voice of happy experience. Digital?
Analog?  Combination? Mac, PC, NeXT, UNIX platforms all OK. Turn key
systems would be great but we can code.  I already have info on
MacSpeechLab/SoundScope from GW and MacRecorder from
Macromind/Paracomp but am not committed to Macs. I'll post a summary
of email responses. Thanks,

- Tom