Voice and Identity: Source, Filter, Biometric

This project compares the performance of different methods for forensic voice comparison – from linguistics and phonetics, acoustics, and automatic speaker recognition – on the same set of recordings.

The ultimate aim is to improve methods in forensic voice comparison, taking a major step towards the development of a methodology that is more transparent, validated, and replicable. This outcome will benefit academics and forensic practitioners, the public, judicial systems, and investigative/security agencies.

Visit the project page