Abstract

Speech intelligibility of mild and moderately hearing-impaired listeners can be successfully accounted for in terms of audibility, as quantified by the articulation index theory. For people with more severe losses, performance is often overestimated by audibility. It is hypothesized that deficits in frequency and temporal resolution may be responsible for the reduced ability of these listeners in using an audible signal for speech intelligibility. This study aimed to measure the amount of information that could be extracted from an audible signal by listeners with different degrees of hearing loss at different frequency regions. Twelve normally hearing and 40 hearing-impaired listeners were tested using filtered sentence material presented at different levels. The results showed that some listeners performed better than normal at low sensation levels but much poorer at high sensation levels. The conventional method of modifying the index by a multiplicative factor would not be appropriate because for many subjects, a smaller factor would be needed at high than at low sensation levels. A modification procedure which incorporates a level-dependent and frequency-dependent correction is proposed, and is verified using speech results obtained from sentences and nonsense syllables tests.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call