Forensic analysis of auditorily similar voices

ABSTRACT Purpose: to verify contributions of acoustic spectrographic analysis in the forensic identification speakers with auditorily similar voices, considering distinctive behavior parameters: formants vowel “é”, connected speech, mean fundamental frequency Hz, linear prediction curve “é” and area; propose an objective method use analyzed parameters. Methods: a quantitative, qualitative descriptive study, conducted Pernambuco on 16 pairs male siblings, aged 18-60 years. The subjects recorded videos from which audios were extracted, numbered sent three examiners, two groups: older brothers younger brothers, for perceptual-auditory pairing. correct pairings, indicated by at least submitted analysis. statistical tests included Wilcoxon, Kruskal-Wallis Bonferroni, p<0.05. Results: results analyses not enough distinguish voices. Unprecedentedly, measurements areas generated graphs, significance was observed. Conclusion: it concluded that, among parameters studied, objectively effectiveness distinguishing