Abstract
The purpose of this study was to investigate the improvement of the speech recognition score (SRS) by amplification at the conversational level, and to compare it with the expected improvement estimated from aided and unaided audibility of the test sound. The monosyllabic SRSs of 589 cases measured at the level of 65dBSPL in a sound field with and without a hearing aid were analyzed. The expected improvement of the SRS was estimated under the assumption that the dependence of both aided and unaided SRS on the speech intelligibility indices (SIIs) of the test sound is the same.
The SRS showed maximum improvement in cases of moderate hearing loss, and the average improvement was 31.7% for the cases with a mean hearing threshold level of between 40 to 90dB. In a multiple regression analysis, the maximum correlation with the improvement of the SRS was observed for 3 independent variables: unaided SIIs, increase in SII by amplification, and the maximum articulation score (MAS) of the unaided ear. The average SRS improvement was the same as the average expected improvement in a group of 373 cases in which the aided MAS was not significantly lower than the unaided MAS. This means that, on average, the quality of the speech sound at the conversational level emanating from the hearing aid did not so deteriorate as to decrease the SRS as compared with that expected from the aided hearing threshold if the MAS did not deteriorate with the use of a hearing aid.
