Speechreading supplemented with frequency‐selective sound‐pressure information

The benefit of supplementing speechreading with frequency‐selective sound‐pressure information was studied by auditorily presenting this information to normal‐hearing listeners. The sound‐pressure levels in one or two frequency bands of the speech signal with center frequencies of 500, 1600, and 3160 Hz, respectively, and with 1‐ or 1/3‐ oct bandwidth were used to amplitude‐modulate pure‐tone carriers with frequencies equal to the center frequencies of the filter bands. Short sentences were presented to 18 normal‐hearing listeners under the conditions of speechreading‐only and speechreading combined with the sound‐pressure information. The mean number of correctly perceived syllables increased from 22.8% for speechreading‐only to 65.7% when sound‐pressure information was supplied in a single 1‐oct band at 500 Hz and to 86.7% with two 1‐oct bands at 500 and 3160 Hz, respectively. The latter signal scored only 26.7% correct syllables without accompanying visual information.