Differences and distinguishability in the acoustic characteristics of hello in voices of similar-sounding speakers

Abstract

Abstract Forensic Phonetics is an important application of Linguistics that has emerged as a discipline over the last decade. This paper describes a Forensic Phonetic experiment which investigates the nature of within- and between-speaker variation in the acoustic characteristics of the word hello in demonstrably similar-sounding voices. The nature of within-segment variation is determined in repeats of the same word said under different prosodic conditions in order to exclude as much of the linguistically determined variation as is consistent with the realities of the forensic situation, thus providing a good estimate of variation associated with speakers. Intonationally varying tokens of the naturally produced single word utterance hello from six adult Australian males are compared with respect to fundamental frequency, and to centre frequencies and bandwidths of the F-pattern below 5 KHz. This comprises the first five formants and extra resonances, including a possible singer’s formant and tracheal resonance. Results show that between-speaker acoustic differences are pervasive, though not ubiquitous. Magnitudes of between-speaker differences are presented for all parameters, and their forensic significance evaluated. ANOVA and chi-square tests show that even similar-sounding voices differ significantly in their acoustics, especially centre frequencies of F2-F4, formant bandwidth, and incidence of extra resonances. Simulated forensic conditions show that some of these differences are not realistically demonstrable. Nevertheless, there remain sufficient significant differences to distinguish 13 out of 15 pairs: a value of 13% for the denominator of the associated Bayesian Likelihood Ratio for the prosecution hypothesis. Directions for future research are indicated.