An investigation into the effect of limiting the frequency bandwidth of speech on speech recognition in adult cochlear implant users

Milchard, Abigail J. and Cullington, Helen E. (2004) An investigation into the effect of limiting the frequency bandwidth of speech on speech recognition in adult cochlear implant users. International Journal of Audiology, 43 (6), 356-362. (doi:10.1080/14992020400050045).

Abstract

The purpose of this study was to investigate the effect of the limited-frequency bandwidth employed by telephones (300-3400 Hz) on speech recognition in adult cochlear implant users. The Four Alternative Auditory Feature (FAAF) test was used in four conditions: unfiltered and in three filtered conditions of 300-4500 Hz, 300-3400 Hz and 300-2500 Hz. Ten subjects implanted with the Nucleus CI24M device and 10 normal-hearing listeners were assessed to examine differences between word discrimination scores in each condition. Scores obtained from the 300-3400-Hz and 300-2500-Hz filtered conditions were significantly worse than those with unfiltered speech for the cochlear implant subjects, decreasing by 17.7% and 21.4%, respectively, from scores with unfiltered speech. By contrast, the normal-hearing listeners did not experience difficulties in discriminating between words in any of the conditions. Analysis of the word errors demonstrated that the reduction in implant subject scores with bandwidth arose from errors in place of articulation. Filtering speech in this way has a significant effect on speech recognition for cochlear implant subjects but not normal-hearing listeners. Hence, the limitations of the normal telephone bandwidth can be expected to have a negative effect on speech recognition for cochlear implant users using the telephone.

This record has no associated files available for download.

Contributors

Author: Helen E. Cullington

Download statistics