Year: 1983
Type: conference-abstract
Abstract: Discrimination between two 300-ms synthesized syllables, /da/ and /ga/, presented in the middle of a 900-ms narrow-band noise, was measured in three normal and six cochlearly impaired listeners. The segmental distinction was conveyed by the frequency of the initial burst and the form of the third-formant transition, which was 50 ms in duration: For /ga/ the transition started around 1.9 kHz and rose to its steady-state value of 2.5 kHz and for /da/ it started around 3.0 kHz and fell to 2.5 kHz. The narrow-band noises were 300 Hz wide and were centered between 1 and 3.5 kHz. For the normal listeners, the speech was set to 70 dB SPL and the noise to 80 dB SPL. For the impaired listeners, the levels were chosen to yield better than 95% discrimination in the easiest condition, but poor discrimination in the most difficult condition. Both groups of listeners showed the poorest performance when the noise was centered at 2.5 kHz. As the noise moved away from this critical region performance improved for both groups, but more rapidly for the normal listeners. These results indicate that the impaired listeners' reduced frequency selectivity may contribute to their difficulty of discriminating speech in noise. [Supported by NSF and NIH.]
DOI: 10.1121/1.2020743
Authors: Søren Buus, Joanne L. Miller, Bertram Scharf
Institution: Northeastern University
Language: English
Subfield: Speech and Hearing
Field: Health Professions
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2051957502