Potential benefits of varying the frequency‐gain characteristic for speech reception in noise

Abstract

Adaptive linear filtering is currently employed in some single‐microphone hearing aids in an attempt to improve speech reception in noise. The potential benefit of this strategy depends on the spectral spread of masking and the degree to which the spread can be reduced by changing the frequency‐gain characteristic. In this study these effects for subjects with normal hearing under static listening conditions are examined. In the unprocessed condition, subjects were presented with nonsense syllables masked by a 96‐dB SPL octave‐band noise centered on a low, mid, or high frequency. A frequency‐gain characteristic was then imposed with the goal of reducing the overall level in the frequency band containing the octave‐band noise. This processing resulted in large increases (about 45 percentage points) in consonant‐correct scores compared to the unprocessed condition for the low‐frequency octave noise band, and smaller increases for the mid‐ and high‐frequency noise conditions. A similar pattern of improvement in speech reception threshold was observed by van Dijkhuizen et al. [J. Acoust. Soc. Am. Suppl. 1 86, S82 (1989)] under comparable stimulus conditions. Results will be discussed in the context of articulation theory, which can serve as a basis for the design of adaptive filtering algorithms. [Work supported by NIH.]