The relative contribution of visual and auditory components of speech to speech intelligibility as a function of three conditions of frequency distortion.

Abstract

Twenty normal-hearing college students were tested to establish speech discrimination scores under three conditions of sensory input and four conditions of frequency filtering. The speech material used comprised Phonetically Balanced Monosyllabic words from CID Auditory Test W-22. The speech sample was presented under the following sensory input conditions: (1) vision alone, (2) audition alone, (3) vision and audition combined. The frequency spectrum of the speech sample was varied by a filter system to produce the following approximate band widths: (1) unfiltered, (2) 400 Hz low pass, (3) 1800 Hz high pass, (4) 400–2200 Hz band pass. The results indicate that the mode of presentation, frequency filter band widths, and interaction of mode of presentation and frequency filter band widths, all effect speech discrimination. Greater dependency on visual clues was evidenced as auditory distortion increased. The data were found to be significant at the 0.01 level of confidence.