Derivative-free optimization of hearing aid parameters

Shu Hsien Chu, Mingyi Hong, Zhi Quan Luo, Kelly Fitz, Martin McKinney, Tao Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Loudness restoration approaches to hearing aid fitting prescribe gain and compression so as to restore the loudness perceived by a hearing-impaired listener to that perceived by a listener with normal-hearing. Restoring the loudness perception to normal is complicated by the spread of excitation at high stimulus levels that causes intense stimuli at low frequencies to be 'heard' and to contribute to the perceived loudness at high frequencies, producing excess loudness growth and poor sound quality. We apply derivative-free optimization algorithms to find a configuration of hearing aid gain and compression parameters that restores specific loudness perception of a hearing impaired listener to that of a normal hearing listener, while simultaneously minimizing the across-frequency spreading of excitation, and ensuring the feasibility of the resulting hearing aid parameters.

Original languageEnglish (US)
Title of host publication2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings
Pages393-397
Number of pages5
DOIs
StatePublished - Oct 18 2013
Event2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Vancouver, BC, Canada
Duration: May 26 2013May 31 2013

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Other

Other2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
Country/TerritoryCanada
CityVancouver, BC
Period5/26/135/31/13

Keywords

  • Hearing aids
  • derivative-free optimization
  • dynamic range compression
  • loudness
  • spread of excitation

Fingerprint

Dive into the research topics of 'Derivative-free optimization of hearing aid parameters'. Together they form a unique fingerprint.

Cite this