[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: standards of speech intelligibilty for the hearing impaired?



Ward:



I have not yet found a clear failure using the audiogram with Fletcher's full-blown AI calculation, so my working hypothesis is that "additional factors" are not necessary. I've examined the Fletcher AI's ability to predict speech intelligibility scores using data from many different published reports, and I'm always looking for more published data which may pose a challenge to this hypothesis. Many (most) published studies provide inadequate descriptions of calibration for the purpose of AI calculation.



Regarding your question, I have looked at AI predictions for hearing-impaired listeners in steady-state noise, and I found that no correction factors were necessary (see Rankovic 1997; 1998 in previous email). I evaluated data from a study of hearing-impaired listeners in which CNC word lists were presented in a multitalker babble taken from the SPIN test, and I found word score predictions were good (I haven't published this finding). It is notable that the SPIN babble was designed to have no substantial temporal holes so it is essentially a steady noise.



Single-talker interference may be an interesting case that I have not yet studied . Also, I have not yet examined the ability of the AI to predict speech intelligibility in background noises constructed to have temporal holes of controlled depth and duration, although I am aware that some researchers have collected this type of data.



Christine Rankovic, PhD

rankovic@xxxxxxxxxxxxxxxx





----- Original Message ----- From: "Ward R. Drennan" <drennan@xxxxxxxxxxxxxxxx>
To: "Christine Rankovic" <rankovic@xxxxxxxxxxxxxxxx>
Cc: <AUDITORY@xxxxxxxxxxxxxxx>
Sent: Thursday, February 21, 2008 1:54 PM
Subject: Re: standards of speech intelligibilty for the hearing impaired?


Sure, the audiogram (ie audibility) is the primary predictor of speech intelligibility, but don't you think the additional factors could improve predictability of practical tasks such as speech intelligibility in noise and babble backgrounds?



Ward R. Drennan, Ph. D.
VM Bloedel Hearing Research Center
Department of Otolaryngology
University of Washington Box 357923
Seattle, WA 98195
Office: (206) 897-1848
Fax: (206) 616-1828