Fusing linguistic and acoustic information for automated forensic speaker comparison
Science & Justice · 2024
DOI: 10.1016/j.scijus.2024.07.001 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science & Justice, 2024. Verfasst von E K Sergidou und 5 weiteren Autor:innen. Beteiligte Einrichtung: Netherlands Forensic Institute, PO Box 24044, 2490 AA The Hague, the Netherlands; University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the N…. Hauptthemen laut MEDLINE: Forensic Sciences. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 85 Literaturangaben. Gefördert von: Horizon 2020 European Innovation Council Fast Track to Innovation, Horizon 2020 Framework Programme.
Abstract (Originalsprache)
Verifying the speaker of a speech fragment can be crucial in attributing a crime to a suspect. The question can be addressed given disputed and reference speech material, adopting the recommended and scientifically accepted likelihood ratio framework for reporting evidential strength in court. In forensic practice, usually, auditory and acoustic analyses are performed to carry out such a verification task considering a diversity of features, such as language competence, pronunciation, or other linguistic features. Automated speaker comparison systems can also be used alongside those manual analyses. State-of-the-art automatic speaker comparison systems are based on deep neural networks that take acoustic features as input. Additional information, though, may be obtained from linguistic analysis. In this paper, we aim to answer if, when and how modern acoustic-based systems can be complemented by an authorship technique based on frequent words, within the likelihood ratio framework. We consider three different approaches to derive a combined likelihood ratio: using a support vector machine algorithm, fitting bivariate normal distributions, and passing the score of the acoustic system as additional input to the frequent-word analysis. We apply our method to the forensically relevant dataset FRIDA and the FISHER corpus, and we explore under which conditions fusion is valuable. We evaluate our results in terms of log likelihood ratio cost (Cllr) and equal error rate (EER). We show that fusion can be beneficial, especially in the case of intercepted phone calls with noise in the background.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Netherlands Forensic Institute, PO Box 24044, 2490 AA The Hague, the Netherlands; University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands. Electronic address: e.sergidou@nfi.nl.
- Netherlands Forensic Institute, PO Box 24044, 2490 AA The Hague, the Netherlands.
- Brno University of Technology, Boˇzetˇechova 2, Brno 61266, Czech Republic.
- University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands.
- Netherlands Forensic Institute, PO Box 24044, 2490 AA The Hague, the Netherlands; University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands.
Förderung
- Horizon 2020 European Innovation Council Fast Track to Innovation
- Horizon 2020 Framework Programme
Zitieren
APA
Sergidou, E. K., Ypma, R., Rohdin, J., Worring, M., Geradts, Z., & Bosma, W. (2024). Fusing linguistic and acoustic information for automated forensic speaker comparison. Science & Justice. https://doi.org/10.1016/j.scijus.2024.07.001
Vancouver
Sergidou EK, Ypma R, Rohdin J, Worring M, Geradts Z, Bosma W. Fusing linguistic and acoustic information for automated forensic speaker comparison. Science & Justice. 2024. doi:10.1016/j.scijus.2024.07.001
BibTeX
@article{Sergidou2024,
author = {Sergidou, E K and Ypma, Rolf and Rohdin, Johan and Worring, Marcel and Geradts, Zeno and Bosma, Wauter},
title = {Fusing linguistic and acoustic information for automated forensic speaker comparison},
journal = {Science & Justice},
year = {2024},
doi = {10.1016/j.scijus.2024.07.001}
}
EUVIMED