Comparing the effects of target and masker voice familiarity on children’s speech-in-speech recognition

Abstract

Familiarity with a target voice has been shown to aid spoken word recognition for both children and adults. Familiarity effects of a target voice are not well understood, and even less is known about the impacts of masker voice familiarity. The present study investigated long-term target and masker voice familiarity effects on children’s speech-in-speech recognition by using the child’s own mother’s voice as stimuli. Open-set sentence recognition thresholds (SRTs) were measured adaptively in a two-talker speech masker. Twenty children were tested in three conditions, hearing their mother’s voice and two unfamiliar female voices: (1) Familiar Target/Unfamiliar Masker; (2) Familiar Masker/Unfamiliar Target; and (3) Unfamiliar Target/Unfamiliar Masker. Condition 1 measured effects of target familiarity; Condition 2 examined effects of masker familiarity; Condition 3 served as baseline. Results showed SRTs were significantly better when target speech was spoken by a familiar talker (the mother). When background speech was spoken by the familiar talker, performance was worse relative to an unfamiliar masker. This suggests that children do benefit from long-term voice familiarity with the target but not the masker speech. Thus, voice familiarity may impact children’s ability to attend to the target speech but does not always improve segregation.