Second Biennial Hearing Aid Research and Development Conference

September 22-24, 1997
National Institutes of Health
Bethesda, Maryland

Speech Recognition in Amplitude-modulated Noise: Effects of Hearing Loss, Simulated Loss, and Amplification

Speech recognition in normal-hearing listeners is often better in an amplitude- modulated noise than in a steady-state noise. This masking release, however, is greatly reduced in hearing-impaired listeners. It is unclear whether this is simply due to reduced audibility. The purpose of the present study was to investigate the role of audibility in this masking release, and to determine whether amplification would increase the masking release for listeners with hearing loss.

Masking release was determined by comparing the signal-to-noise ratio for 50% correct sentence recognition in a background of speech-shaped noise that was unmodulated or modulated by a 10-Hz square wave. The results from the 12 hearing-impaired listeners were compared with those from 16 normal- hearing listeners who listened under various filtering conditions meant to simulate different hearing losses. The hearing-impaired listeners were also tested with amplification.

The results showed 13 to 23 dB of masking release for the normal-hearing listeners (unfiltered) and 0 to 10 dB of masking release for the hearing- impaired listeners (unaided). Generally, the masking release was greater for those listeners with better quiet thresholds. When tested with amplification, most hearing-impaired listeners showed increased masking release. In conditions of equal audibility, approximately half of the hearing-impaired listeners achieved the same (within 3 dB) masking release as the normal- hearing listeners. This result suggests that audibility is the main factor in the reduced masking release for some hearing-impaired listeners. These results could have implications for determining amplification characteristics of a hearing aid.

Return to Conference Abstracts List

Top