Speechreading supplemented with auditorily presented speech elements in the profoundly hearing impaired.

For many profoundly hearing-impaired listeners (hearing loss > 90 dB HL) speechreading is the most important means of communication; amplified speech may provide, at best, additional information to speechreading. In order to improve audiovisual communication, three speech pattern elements comprising voice-fundamental frequency (f0), the first formant (F1), and the first and the second formant (F1F2) were presented as supplements to speechreading. A fourth condition consisted of a natural speech supplement, a fifth of speechreading only. Twenty subjects were tested; all audiovisual speech scores were significantly higher than the purely visual scores. Audiovisual scores for amplified, natural speech were significantly higher than those for f0 and F1F2 coded speech. Scores for natural speech and for F1 coded speech were not significantly different. The relations between the increase in audiovisual speech scores over the visual scores and measures of difference limen for frequency (DLf) and gap detection were not clear. The most prominent correlations with the speech scores were found for the DLf at 125 Hz and for gap detection.