Re: Stop consonant identification based on initial spectra?

[Alex Francis <francisa@xxxxxxxxxx>]

Peter,

It does appear that the place-of-articulation related patterns of stop
consonant burst spectra are relatively invariant:

Stevens, K. N., & Blumstein, S. E. (1981).  The search for invariant
acoustic correlates of phonetic features.  In Eimas, P. D., & Miller, J. L.
(Eds.) Perspectives on the Study of Speech (pp. 1-38). Lawrence Erlbaum
Associates, Inc.

Furthermore, burst spectra are apparently usable as acoustic cues to place
of articulation, but they do not appear to be as useful as formant
transitions, and given a choice listeners tend to prefer formant transitions:

Walley, A. C., & Carrell, T. D. (1983).  Onset spectra and formant
transitions in the adult's and child's perception of place of articulation
in stop consonants.  Journal of the Acoustical Society of America, 73,
1011-1022.

It's even difficult (though probably not impossible) to train listeners to
use burst properties instead of formant transitions. It seems to be
comparatively easier to train listeners to use formant transitions instead
of bursts.

Francis, A. L., Baldwin, K., & Nusbaum, H. C. (2000).  Effects of training
on attention to acoustic cues. Perception and Psychophysics,62(8), 1668-1680.

Where did you see the poster?  Do you remember the authors' names?

-alex


At 11:06 AM 3/2/2005, Marvit, Peter wrote:
> I'm sorry to interrupt the current frenzy of pet anecdotes (in which no one
> has yet mentioned fish)...
>
> I'm looking for a reference that reports whether or not humans can identify
> stop consonants based on their initial spectra--before the formant
> transitions to the following vowel. Secondarily (though I suppose more
> fundamentally), are the initial spectra (first 10 msec or however long
> *before* formant transitions) invariant with respect to following vowels?
> Differences between voiced and unvoiced?
>
> Background: I had been well indoctrinated in the motor theory of speech
> perception, teaching my students the wonders of categorical perception of
> stop consonants despite widely varying formant transition profiles across
> different vowels (i.e., /di/ looks rather different than /du/ but we
> identify /d/ in both). A recent conference poster looking at
> neurophysiological spectral representation in non-human primate suggested
> that response to spectra of stop consonants (without the following formant
> transitions) was sufficient to distinguish and identify them. Alas, I did
> not get the relevant human reference and have been unable to find one in an
> informal search of my reference books and MEDLINE.
>
> Thanks in advance,
> Peter
>
> : Peter Marvit, PhD                          <pmarvit@xxxxxxxxxxxxxxxxx> :
> : Dept. Anatomy and Neurobiology    University of Maryland Medical School:
> : 20 Penn Street, HSF II, Room S251                  Baltimore, MD 21201 :
> : phone 410-706-1272      http://www.theearlab.org      fax 410-706-2512 :




  Alexander L. Francis        http://web.ics.purdue.edu/~francisa
  Assistant Professor                          francisA@xxxxxxxxxx
  Speech, Language & Hearing Sciences       ofc. +1 (765) 494-3815
  Purdue University                         lab. +1 (765) 494-7553
  500 Oval Drive                            fax. +1 (765) 494-0771
  West Lafayette IN 47907-2038
  USA
------------my employer requires the following addendum------------
CONFIDENTIALITY NOTICE: This email and the documents accompanying
this email may contain legally privileged confidential information.
The information is intended only for the use of the individual or
entity named as recipient. If you are not the intended recipient,
you are hereby notified that any disclosure, reproduction or
distribution of this email or its attachments, or the taking of any
action in reliance on the contents of this email or its attachments,
is strictly prohibited. If you have received this email in error,
please notify the sender by return email.