SPEECHREADING ENHANCEMENT USING A SINUSOIDAL SUBSTITUTE FOR VOICE FUNDAMENTAL-FREQUENCY

This study compared the effectiveness of three acoustic supplements to speechreading: the low-pass-filtered output of an elecrtroglottograph , a variable-frequency sinusoidal substitute for voice fundamental fre quency (F0), and a constant-frequency sinusoidal substitute that serve d as a representation of voicing. Both sinusoidal signals were synthes ized at constant amplitude during periods of voicing. The sinusoidal s ignals were prepared off-line by a combination of automatic and manual estimation of the F0 contours of video-recorded sentences. These sign als were then resynchronized with the audio portions of the original r ecording. In 12 normally-hearing adults, the electroglottograph signal and the variable-frequency sinusoidal F0 substitute both increased th e number of words recognized in sentences of known topic by between 30 and 35 percentage points. The magnitude of this effect was greater fo r longer sentences but independent of basic speechreading ability. The constant-frequency substitute provided a 13 percentage point increase , suggesting that approximately one-third of the F0 speechreading enha ncement effect could be accounted for by voicing detection alone.