Abstract
Formant characteristics are most commonly part of forensic speaker comparison (FSC). However, only formants F1 to F3 typically occur in evidence material because it is mostly recorded via telephone. Given recent technological advances in telephony (e.g. WeChat or WhatsApp) higher formants (F4-F5) are becoming increasingly part of evidence material. The present study investigated the speaker- distinguishing properties of F1 to F5 of three sustained vowels /i/, /y/ and /ɤ/ in Mandarin produced by 20 young male speakers. Based on discriminant analysis, for each single formant, the best predictors were F5 for /i/ and F4 for /y/ and /ɤ/. Classification performance varied between vowels. Inclusion of two and three formants yielded higher classification rates of 30−80%. The best value was provided by the combination of F2, F4 and F5 of /ɤ/. The value and limitations of F4 and F5 for FSC are discussed.
Cao, Honglin