Grange, Jacques ![]() ![]() ![]() |
Preview |
PDF
- Published Version
Available under License Creative Commons Attribution. Download (1MB) | Preview |
Abstract
Speech reception thresholds (SRTs) in noise improve when the speech and noise sources are spatially separated. This spatial release from masking (SRM) is usually investigated in fixed-head situations. We studied free-head situations in audio and audio-visual conditions. We compared normally- hearing and cochlear-implant (CI) users’ spontaneous and directed head- orientation strategies when attending to speech in noise with a progressively declining signal-to-noise ratio. SRM-model predictions suggested benefits of head orientation away from the target speech that we hypothesized would motivate head rotation. As signal-to-noise ratio declined, observed head tracks differed greatly between listeners. Audio-visual presentation reduced the amount of head rotation. When directed, listeners made more effective use of head rotation. Audio and audio-visual SRTs were acquired at fixed, 0, and 30 deg head orientations with respect to the target speech. At the most favourable 30-deg head orientation, SRM reached 8 and 6 dB for NH listeners and CI users respectively. Lip-reading yielded improvements of 3 and 5 dB on average across conditions. CI users confirmed that training in optimizing both their position and head orientation with respect to target speaker and noise source position in a social setting was both currently missing and likely valuable.
Item Type: | Conference or Workshop Item (Paper) |
---|---|
Date Type: | Publication |
Status: | Published |
Schools: | Psychology |
Subjects: | B Philosophy. Psychology. Religion > BF Psychology |
Publisher: | ISAAR |
ISBN: | 9788799001347 |
ISSN: | 2596-5522 |
Funders: | Action on Hearing Loss |
Date of First Compliant Deposit: | 30 March 2016 |
Date of Acceptance: | 10 January 2014 |
Last Modified: | 28 Oct 2022 09:07 |
URI: | https://orca.cardiff.ac.uk/id/eprint/73383 |
Actions (repository staff only)
![]() |
Edit Item |