Re: Rationale for Critical Bands

[Christine Rankovic <rankovic@xxxxxxxxxxxxxxxx>]

Dear List:

Regarding Brian's comment on speech intelligibility at high intensities:

In his version of the articulation index, Fletcher did not increase the
width of the auditory filter to account for the reduction in speech
intelligibility in quiet observed at high intensity levels, a phenomenon
often referred to as "rollover."  Fletcher attributed rollover to a
combination of self-masking of speech by both simultaneous (masking spread)
and forward (within band) masking, and to ear-generated distortion products
which may arise in conditions such as, for example, under high-frequency
amplification.  Fletcher incorporated methods for calculation these effects
into his definitive 1950 version of the articulation index calculation
published with R. Galt in JASA (this is the same calculation as in
Fletcher's 1953 book).

Fletcher was the first to describe the critical band concept and determine
the bandwidths experimentally, so I'm convinced that his decision not to
broaden the auditory filter was an informed one.  In fact, he oversaw Bell
Labs researchers French and Steinberg, whose 1947 version of the
articulation index (a simplified AI calculation developed for military
applications) included a factor, m, that effectively broadened the filters
to account for excess masking at high intensities.

References:  

H. Fletcher, Speech in Hearing in Communication (1953). Van Nostrand, New
York. (or ASA edition available at asa.aip.org).

H. Fletcher, and R.H. Galt (1950).  Journal of the Acoustical Society of
America vol. 22, 89-151. 

H. Fletcher and W.A. Munson (1933).  Journal of the Acoustical Society of
America vol. 5, 82-108 for original critical band discussion.

N.R. French and J.C. Steinberg (1947).  Journal of the Acoustical Society of
America vol. 19, 90-119.

Christine Rankovic, PhD
Research Scientist
Articulation Inc
94 Berkshire Road
Newton, MA  02460
USA

-----Original Message-----
From: AUDITORY - Research in Auditory Perception
[mailto:AUDITORY@xxxxxxxxxxxxxxx] On Behalf Of Brian C. J. Moore
Sent: Tuesday, June 15, 2010 10:42 AM
To: AUDITORY@xxxxxxxxxxxxxxx
Subject: Re: Rationale for Critical Bands

Dear All

At 15:00 15/06/2010, Wiebe Horst wrote:
>Dear all,
>
>I am getting more and more confused by this discussion.
>It would help if every contributor started by giving his/her
>definition of Critical Band.
>I am used to using Scharf's (1970) definition:
>"The critical band is a band of that width at which subjective
>responses  change rather abruptly."

The problem with this definition is that there are few if any 
auditory responses which "change rather abruptly" at a certain 
bandwidth.  The perceptual changes are nearly always smooth and 
progressive, consistent with auditory filters with rounded tops.

And speech perception does deteriorate at high levels, probably due 
to broadening of the auditory filters.  See, for example:

Studebaker, G. A., Sherbecoe, R. L., McDaniel, D. M., Gwaltney, C. 
A., 1999. Monosyllabic word recognition at higher-than-normal speech 
and noise levels. J. Acoust. Soc. Am. 105, 2431-2444.

Best wishes,

Brian Moore

Brian C. J. Moore, Ph.D, FMedSci, FRS,
Professor of Auditory Perception,
Department of Experimental Psychology,
University of Cambridge,
Downing Street,
Cambridge CB2 3EB,
UK
Tel. +44 (0) 1223 333574
Fax. +44 (0) 1223 333564
http://hearing.psychol.cam.ac.uk