Answers are grounded in this paper and cite their sources.
The perception of voice similarity was examined in 5-year-old children with normal hearing sensitivity and in pediatric cochlear implant users, 5-12 years of age. Recorded sentences were manipulated to form a continuum of similar-sounding voices. An adaptive procedure was then used to determine how acoustically different, in terms of average fundamental and formant frequencies, 2 sentences needed to be for a child to categorize the sentences as spoken by 2 different talkers. The average spectral characteristics of 2 utterances (including their fundamental frequencies) needed to differ by at least 11%-16% (2-2.5 semitones) for normal-hearing children to perceive the voices as belonging to different talkers. Introducing differences in the linguistic content of the 2 sentences to be compared did not change performance. Although several children with cochlear implants performed similarly to normal-hearing children, most found the task very difficult. Pediatric cochlear implant users who scored above the group mean of 64% of words correct on a monosyllabic open-set word identification task categorized the voices more like children with normal hearing sensitivity.
No takes yet. Share an insight, caveat, or question.
Cleary et al. (2005) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: