The effects of visual-only vowel identification training on speech recognition for adults with hearing loss

Abstract

Speechreading, the use of visual cues to speech, may provide benefit for listeners with hearing loss in everyday communication beyond that achieved with hearing aids alone. Recent work examined the effects of a novel approach to speechreading training, for normal-hearing listeners tested in masking noise designed to simulate a hearing loss [C. Richie and D. Kewley-Port, J. Acoust. Soc. Am. 117, 2570 (2005)]. That study showed improvements in sentence-level speech recognition for listeners trained to speechread vowels under audiovisual conditions, compared to listeners trained to speechread consonants and untrained listeners. The present study examined the effects of vowel identification speechreading training on sentence-level speech recognition, for adults with hearing loss. Computer-based training and testing was administered visual-only, in order to determine if training in the visual modality alone leads to gains in speech recognition. Results showed that vowel identification training under visual-only conditions led to improved vowel recognition, but that improved vowel identification did not generalize to increased sentence-level speech recognition. These findings suggest that some acoustic information, however minimal, may be beneficial during speechreading training and required for improved sentence-level speech recognition by adult listeners with hearing loss. [Work supported by Butler University HAC Grant No. 027096.]