Scarlett Vanroose-Peer

PhD student

Research Overview

Data contamination is a widely acknowledged phenomenon across many forensic disciplines. However, the forensic speech science community has so far overlooked the dangers surrounding contaminated speech samples. My research investigates the risk and impact of data contamination in forensic voice identification, with the implications of improving trust within the criminal justice system as well as intelligence and security operations.

AI Investigators: Spot the Bot
Types of Public engagement and outreach - Festival/Exhibition

VoiceID 2026
Participation in conference - Academic

International Association for Forensic Phonetics and Acoustics 2026
Participation in conference - Academic

The risk of data contamination in (forensic) voice identification: a perception experiment using voice-mixed speech samples
Invited talk

Best Student Poster Award
Prize (including medals and awards)