Reed C M, Durlach N I, Braida L D, Schultz M C
J Speech Hear Res. 1982 Mar;25(1):108-16. doi: 10.1044/jshr.2501.108.
Performance on identifying consonant and vowel stimuli through the Tadoma methods of speechreading was studied for a deaf-blind, experienced user of Tadoma. For a set of 24 consonants in CV syllables, performance averaged 55%. For a set of 15 vowels and diphthongs in CVC syllables, performance averaged 56%. Confusion matrices derived from the identification tests were analyzed for performance on a set of articulatory/phonological features, the results of the feature analysis for consonants indicated that the features voicing, lip rounding frication, and place were well-perceived; for vowels, and features lip rounding, tenseness, and vertical lip separation were well-perceived.
