Evaluation of a non-linear spectral subtraction noise suppression scheme in cochlear implant users

Carl Verschuur, Mark Lutman, Nor Haniza Abdul Wahat

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

The aim of the study was to determine benefit to speech recognition in noise by adult cochlear implant users with the non-linear spectral subtraction (NSS) noise suppression strategy. Users of the Nucleus 22 or Nucleus 24 cochlear implant systems were tested with sentence materials combined with stationary noise at +5 and +10dB signal to noise ratio (SNR), with and without NSS processing applied offline. Sentence scores were significantly higher with NSS processing, for both SNRs. The effect was greater at +5 dB SNR (12% improvement with NSS) than at +10dB SNR (5% improvement with NSS). These results are promising and suggest that online implementation of NSS as part of cochlear implant processors has the potential to yield benefits for speech recognition in noise.

Original languageEnglish
Pages (from-to)188-193
Number of pages6
JournalCochlear Implants International
Volume7
Issue number4
DOIs
Publication statusPublished - 2006
Externally publishedYes

Fingerprint

Cochlear Implants
Noise
Signal-To-Noise Ratio

Keywords

  • Algorithms
  • Cochlear implant
  • Noise
  • Speech perception

ASJC Scopus subject areas

  • Otorhinolaryngology

Cite this

Evaluation of a non-linear spectral subtraction noise suppression scheme in cochlear implant users. / Verschuur, Carl; Lutman, Mark; Abdul Wahat, Nor Haniza.

In: Cochlear Implants International, Vol. 7, No. 4, 2006, p. 188-193.

Research output: Contribution to journalArticle

@article{e883141354a0409d8ccd6af08478254d,
title = "Evaluation of a non-linear spectral subtraction noise suppression scheme in cochlear implant users",
abstract = "The aim of the study was to determine benefit to speech recognition in noise by adult cochlear implant users with the non-linear spectral subtraction (NSS) noise suppression strategy. Users of the Nucleus 22 or Nucleus 24 cochlear implant systems were tested with sentence materials combined with stationary noise at +5 and +10dB signal to noise ratio (SNR), with and without NSS processing applied offline. Sentence scores were significantly higher with NSS processing, for both SNRs. The effect was greater at +5 dB SNR (12{\%} improvement with NSS) than at +10dB SNR (5{\%} improvement with NSS). These results are promising and suggest that online implementation of NSS as part of cochlear implant processors has the potential to yield benefits for speech recognition in noise.",
keywords = "Algorithms, Cochlear implant, Noise, Speech perception",
author = "Carl Verschuur and Mark Lutman and {Abdul Wahat}, {Nor Haniza}",
year = "2006",
doi = "10.1002/cii.318",
language = "English",
volume = "7",
pages = "188--193",
journal = "Cochlear Implants International",
issn = "1467-0100",
publisher = "Maney Publishing",
number = "4",

}

TY - JOUR

T1 - Evaluation of a non-linear spectral subtraction noise suppression scheme in cochlear implant users

AU - Verschuur, Carl

AU - Lutman, Mark

AU - Abdul Wahat, Nor Haniza

PY - 2006

Y1 - 2006

N2 - The aim of the study was to determine benefit to speech recognition in noise by adult cochlear implant users with the non-linear spectral subtraction (NSS) noise suppression strategy. Users of the Nucleus 22 or Nucleus 24 cochlear implant systems were tested with sentence materials combined with stationary noise at +5 and +10dB signal to noise ratio (SNR), with and without NSS processing applied offline. Sentence scores were significantly higher with NSS processing, for both SNRs. The effect was greater at +5 dB SNR (12% improvement with NSS) than at +10dB SNR (5% improvement with NSS). These results are promising and suggest that online implementation of NSS as part of cochlear implant processors has the potential to yield benefits for speech recognition in noise.

AB - The aim of the study was to determine benefit to speech recognition in noise by adult cochlear implant users with the non-linear spectral subtraction (NSS) noise suppression strategy. Users of the Nucleus 22 or Nucleus 24 cochlear implant systems were tested with sentence materials combined with stationary noise at +5 and +10dB signal to noise ratio (SNR), with and without NSS processing applied offline. Sentence scores were significantly higher with NSS processing, for both SNRs. The effect was greater at +5 dB SNR (12% improvement with NSS) than at +10dB SNR (5% improvement with NSS). These results are promising and suggest that online implementation of NSS as part of cochlear implant processors has the potential to yield benefits for speech recognition in noise.

KW - Algorithms

KW - Cochlear implant

KW - Noise

KW - Speech perception

UR - http://www.scopus.com/inward/record.url?scp=34147115271&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34147115271&partnerID=8YFLogxK

U2 - 10.1002/cii.318

DO - 10.1002/cii.318

M3 - Article

VL - 7

SP - 188

EP - 193

JO - Cochlear Implants International

JF - Cochlear Implants International

SN - 1467-0100

IS - 4

ER -