The role of extended high frequencies in children's speech-in-speech recognition

Abstract

This study examined the extent to which school-aged children (5–17 yrs) are able to take advantage of extended high frequency (EHF) energy when recognizing sentences in a two-talker masker. Recent work demonstrated that EHFs can be beneficial for adults' speech-in-speech recognition, but its role in children's speech recognition has not been studied. Given that children have superior EHF hearing compared to adults and that EHF audibility may play an important role in language development, we hypothesized that EHF may be especially useful to children in multitalker contexts. The present study measured children's open-set sentence recognition in a two-talker masker using two filtering conditions: full band versus all stimuli low-pass filtered at 8 kHz. Given that EHF energy emission in speech is highly dependent on the head orientation of the talker, two masker head rotation conditions were tested: both maskers at 45 deg or at 60 deg rotation, relative to the target talker. Preliminary results demonstrate children performed best when EHF was present, indicating they use EHF for speech-in-speech recognition. However, thresholds remained elevated compared to adults, suggesting that while EHF is a salient cue for children, their increased EHF hearing sensitivity (relative to adults) did not increase the EHF benefit.