Machine learning and the auditory nerve

Authors

  • Jeff Bondy Dept. of Electrical Engineering, McMaster University, 1280 Main St. W., Hamilton, Ont. L8S 4K1, Canada
  • Ian C. Bruce Dept. of Electrical Engineering, McMaster University, 1280 Main St. W., Hamilton, Ont. L8S 4K1, Canada
  • Sue Becker Dept. of Electrical Engineering, McMaster University, 1280 Main St. W., Hamilton, Ont. L8S 4K1, Canada
  • Simon Haykin Dept. of Electrical Engineering, McMaster University, 1280 Main St. W., Hamilton, Ont. L8S 4K1, Canada

Keywords:

Algorithms, Hearing aids, Neurology, Optimization, Parameter estimation, Problem solving, Signal encoding, Auditory nerves, Auditory systems, Neural Articulation Index (NAI), Neurocompensators, Speech Transmission Index (STI)

Abstract

The application of machine learning to the auditory system is studied. The application consists of four models such as model of the normal auditory system, impaired auditory system, processing block to train, and an error metric. The statistical differences between the normal and impaired auditory nerve reponses shows a loss of contrast between different auditory landmarks. It is expected that better segmentation will lead to more normal streaming, allowing the hearing-aid user the ability to unmask spectrally and temporally and also a normal hearing person.

Downloads

Published

2004-09-01

How to Cite

1.
Bondy J, Bruce IC, Becker S, Haykin S. Machine learning and the auditory nerve. Canadian Acoustics [Internet]. 2004 Sep. 1 [cited 2021 Oct. 25];32(3):64-5. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/1629

Issue

Section

Proceedings of the Acoustics Week in Canada