Re: Consonant speech database

[yund <yund@xxxxxxxxx>]

I have live voice recording of CUNY nonsense syllables that we made
at two different times.  There are male- and female-voice sets (six
exemplars of each syllable) recorded about 15 years ago, 12-bit direct-
to-digital 20-kHZ recordings. There is also another male and female voice
we recorded (6 of each, again) just 2 years ago.  The first two are the
syllables used in my lab in studies of hearing aid signal processing,
primarily multichannel compression systems, published since 1990.  We
did our own recordings because we could not isolate the syllables from
the carrier phrases in the standard CUNY NST recordings -- and we could
not use the full carrier phrase and syllables.  I have not made these
stimuli generally available, but I could put a zipped copy on the web
or attach it to an email, if you or anyone else would like a copy.  I
have not taken the vast amounts of normative data that should be available
for a standard set of stimuli, and that, plus the fact that it is only a
isolated-syllable version of the CUNY NST, is why I have not made it generally
available in the past.

Bill Yund

Gopikrishna Allu wrote:
> Dear Members,
>
> I am working on a project where I need to do some recognition experiments only
> on consonant sounds (or in other words, speech dominated by consonant sounds).
> Can anyone please suggest me where I can find such a Consonant speech database
> or Consonant-Vowel-Consonant (CVC) speech database?
>
> Thank you very much.
>
> Gopi Allu,
> University of Rhode Island.


--
E. William Yund, Ph.D.

Hearing Loss Research Laboratory
VA Medical Center
150 Muir Road
Martinez, CA 94553

yund@xxxxxxxxx
(925)372-2296
FAX (925)228-5738