Effects of the fitting parameters of a two-channel compression system on the intelligibility of speech in quiet and in noise

Abstract

These experiments were carried out to assess how accurately the gains and compression ratios in a two-channel compression system needed to be set. We used as a research tool a laboratory version of a two-channel full-dynamic-range compression system. The system was initially adjusted to suit each hearing-impaired subject according to the manufacturer's recommendations. Then, further adjustments were made to ensure that speech stimuli were both audible and comfortable over a wide range of sound levels. Finally, the settings of the gains and compression ratios were systematically varied from the adjusted values and the effects of this on the intelligibility of speech in quiet and in noise (12-talker babble, levels of 65 and 75 dB SPL) were measured. The results indicated that speech reception thresholds (SRTs) in quiet were significantly adversely affected by decreases in low-level gain. However, SRTs in noise were relatively unaffected by changes in low-level gain. An exception occurred at the higher noise level used, where increases in the low-level gains (with corresponding increases in compression ratios) had a significant adverse effect on the SRTs. It is concluded that, provided excessive low-level gains (associated with high compression ratios) are avoided, the main criteria for fitting such a system should be listening comfort (i.e. achieving an acceptable tonal balance, and avoiding uncomfortably loud sounds) and an appropriate value of the threshold for detecting speech in quiet (which should be a little below 50 dB SPL).

Intra- and Intersubject Variability in Audiometric Measures and Loudness Judgments in Older Listeners with Normal Hearing

“…Numerous investigators have advocated using loudness scaling techniques for fitting hearing aids, especially for fitting programmable multichannel nonlinear hearing aids. 20,21 One popular method of assessing a patient's loudness perception has been to use categorical loudness judgments. For example, Allen and colleagues assessed loudness growth in normal and hearing-impaired subjects using a categorical rating method known as loudness growth in ½-octave bands, or LGOB.…”

Long-term Test-Retest Reliability of Category Loudness Scaling in Normal-Hearing Subjects Using Pure-tone Stimuli

“…Procedures based on investigations by Pascoe (1988) and Allen et al (1990) for fitting of this new generation of hearing aids have been proposed among others by Kiessling et al (1996) and Moore et al (1992). It is well known that obtained loudness levels are highly dependent on the instruction and procedure (see for instance Ritter et al, 1979;Geller & Margolis, 1984;Skinner, 1988;Israelsson et al, 1995).…”

Aided Loudness Growth and Satisfaction with Everyday Loudness Perception in Compression Hearing Aid Users

“…However, little evidence has been found indicating that loudness normalization increases satisfaction in hearing aid users (e.g., Byrne, 1996;Dillon, 1996;Kuk, 1996). In cases of severe recruitment, high compression ratios would be needed to normalize loudness growth, but such ratios could be detrimental to speech perception (e.g., Moore et al, 1992;Neuman et al, 1994;Plomp, 1994;Van Tasell and Trine, 1996;Verschuure et al, 1996). Therefore, the question is raised: Is loudness normalization necessary for listeners to be satisfied with aided loudness (Byrne, 1996)?…”