Speech perception with a cochlear implant sound processor incorporating loudness models

Recently, a new paradigm for processing sounds for cochlear implants has been described that aims to normalize loudness for complex signals, relative to normal acoustic loudness perception [McDermott et al., J. Acoust. Soc. Am. 114, 2190–2197 (2003)]. A first implementation of a sound processor based on this paradigm, called SpeL, has been evaluated with speech perception tests conducted in quiet and in noise. For the five participating subjects, the average scores from these tests for SpeL were very close to those for the conventional ACE scheme. However, audibility was improved by about 5 dB for SpeL compared with ACE.