Introduction
Hearing
loss not only leads to restricted speech information input and
understanding but also can alter speech
output.1Speech intelligibility is an important aspect of speech-language output
that allows a listener to understand what a speaker is
saying.
Therefore, speech intelligibility has even been considered to be the
single most important practical index of oral language abilities for
persons with hearing impairment.2 This positive
outlook has been bolstered by sizable gains in speech intelligibility
after children with hearing loss receive sensory aids at relatively
young ages.3,4
Traditional speech intelligibility assessments are solely based on
subjective judgments.5 Therefore, the accuracy of this
assessment will strongly rely on the experience of each rater, which
could affect the accuracy of evaluations.
An automatic speech recognition
(ASR) system could be used to predict speech intelligibility. The aim of
this study was to compare the efficacy of an
automatic
speech recognition (ASR) system and the traditional method for
evaluating speech intelligibility among children with hearing loss.