Formant dynamics and durations of um improve the performance of automatic speaker recognition systems

摘要

We assess the potential improvement in the performance of MFCC-based automatic speaker recognition (ASR) systems with the inclusion of linguistic-phonetic information. Likelihood ratios were computed using MFCCs and the formant trajectories and durations of the hesitation marker um, extracted from recordings of male standard southern British English speakers. Testing was run over 20 replications using randomised sets of speakers. System validity (EER and Cllr) was found to improve with the inclusion of um relative to the baseline ASR across all 20 replications. These results offer support for the growing integration of automatic and linguistic-phonetic methods in forensic voice comparison.