Skip to content

CoNNear: A CNN model of the human auditory periphery

Latest
Compare
Choose a tag to compare
@HearingTechnology HearingTechnology released this 14 Jul 16:04
· 5 commits to master since this release

Release v1.0 of CoNNearperiphery, a convolutional-neural-network (CNN) model of the human auditory periphery. CoNNearperiphery consists of CoNNearcochlea and CoNNearIHC-ANF and can be used to simulate the outputs of the cochlear, inner-hair-cell and auditory-nerve stages of the human auditory periphery to an audio stimulus. A faster implementation of the CoNNear periphery model can be found under branched, which was presented in the supporting paper "A Neural-Network Framework for the Design of Individualised Hearing-Loss Compensation" by Fotios Drakopoulos and Sarah Verhulst.