| Literature DB >> 19363722 |
Patrick J Boyle1, Andreas Büchner, Michael A Stone, Thomas Lenarz, Brian C J Moore.
Abstract
Cochlear implants usually employ an automatic gain control (AGC) system as a first stage of processing. AGC1 was a fast-acting (syllabic) compressor. AGC2 was a dual-time-constant system; it usually performed as a slow-acting compressor, but incorporated an additional fast-acting system to provide protection from sudden increases in sound level. Six experienced cochlear-implant users were tested in a counterbalanced order, receiving one-month of experience with a given AGC type before switching to the other type. Performance was evaluated shortly after provision of a given AGC type and after one-month of experience with that AGC type. Questionnaires, mainly relating to listening in quiet situations, did not reveal significant differences between the two AGC types. However, fixed-level and roving-level tests of sentence identification in noise both revealed significantly better performance for AGC2. It is suggested that the poorer performance for AGC1 occurred because AGC1 introduced cross-modulation between the target speech and background noise, which made perceptual separation of the target and background more difficult.Entities:
Mesh:
Year: 2009 PMID: 19363722 DOI: 10.1080/14992020802581982
Source DB: PubMed Journal: Int J Audiol ISSN: 1499-2027 Impact factor: 2.117