Does an infant-directed speaking style aid in the separation of different streams of speech?

Abstract

This work explores whether an infant-directed speaking style (IDS) may be easier to separate from background speech noise than is adult-directed speech (ADS). Many of the acoustic cues found in infant-directed speech are similar to ones shown to be important in adult stream segregation (such as differences in voice pitch and pitch variability). In addition, using an infant-directed speech style may serve to make the talker’s voice more dissimilar from any background speech (which is likely to be adult-directed). We explored this issue in three different ways: by examining adults’ ability to separate two different streams of speech varying in register, by examining whether adult speakers use IDS to a greater extent when in the context of noise, and by examining whether infants’ preference for IDS over ADS might be greater in the context of background noise. Implications of these three lines of work will be discussed. [Work supported by NSF and NICHD.]