Lip-Reading Aids Word Recognition Most in Moderate Noise: A Bayesian Explanation Using High-Dimensional Feature Space

Wei Ji Ma, Xiang Zhou, Lars A. Ross et al. · 2009

Read Lip-Reading Aids Word Recognition Most in Moderate Noise: A Bayesian Explanation Using High-Dimensional Feature Space by Wei Ji Ma, Xiang Zhou, Lars A. Ross et al. (2009) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0004638.txt.

Description

Seeing a speaker's lips helps most when noise is moderate, not extreme, a Bayesian model by Wei Ji Ma, Xiang Zhou, Lars Ross, and co-authors explains in PLOS ONE in 2009. Watching a speaker's facial movements can dramatically enhance our ability to comprehend words, especially in noisy environments. A general doctrine of combining senses predicts the benefit should be biggest when hearing is worst.

Reality is different: lip-reading aids word recognition most in moderate noise. The authors explain that peak using a Bayesian model with a high-dimensional feature space. The brain, on this account, weighs each sense by its reliability in a way that produces the hump. The result refines theories of multisensory integration used in hearing aids and speech technology.

Catalog record

CREATORWei Ji Ma, Xiang Zhou, Lars A. Ross et al.
YEAR2009
COLLECTIONPAPERS
FORMATTXT
FILE SIZE76 KB
LICENSECreative Commons Attribution License
RIGHTSCC BY
SOURCEPLOS ONE

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0004638

Browse the full library · About the Tower of Babel