ISCA Archive odyssey 2024
ISCA Archive odyssey 2024

ROXSD: The ROXANNE Multimodal and Simulated Dataset for Advancing Criminal Investigations

Petr Motlicek, Erinc Dikici, Srikanth Madikeri, Pradeep Rangappa, Miroslav Jánošík, Gerhard Backfried, Dorothea Thomas-Aniola, Maximilian Schürz, Johan Rohdin, Petr Schwarz, Marek Kováč, Květoslav Malý, Dominik Boboš, Mathias Leibiger, Costas Kalogiros, Andreas Alexopoulos, Daniel Kudenko, Zahra Ahmadi, Hoang H. Nguyen, Aravind Krishnan, Dawei Zhu, Dietrich Klakow, Maria Jofre, Francesco Calderoni, Denis Marraud, Nikolaos Koutras, Nikos Nikolau, Christiana Aposkiti, Panagiotis Douris, Konstantinos Gkountas, Eleni Sergidou, Wauter Bosma, Joshua Hughes, Hellenic Police Team

The ROXANNE project, conducted under the European Union’s Horizon 2020 Programme, aimed to revolutionize criminal investigations by integrating speech, language, and video technologies with criminal network analysis. Despite the success in technology development, the project faced evaluation challenges due to the scarcity and legal restrictions surrounding real-world criminal activity datasets. In response, we introduce ROXSD, a simulated dataset of communication in organized crime. ROXSD is a set of wiretapped conversations (collected through communication service providers) between drug dealing suspects, following a realistic screenplay (incl. realistic conditions and constraints of a real investigation) prepared by Law Enforcement Agencies (LEAs). With a focus on multimodality and multilinguality, the dataset comprises 20 hours of telephone and video conversations involving 104 speakers, and is further aligned with ground-truth annotations for each modality involved, enabling precise evaluation and development of technologies. In addition, the multimodal data are enhanced with metadata and prior knowledge (e.g., suspects’ biometric profiles) which is typically available as a result of lawfully intercepted communication. This paper introduces ROXSD as a pivotal resource for advancing technology in criminal research (specifically in domain of speech, text and network analysis). ROXSD not only facilitates in the domain of technology development and evaluation but also showcases the potential of simulated datasets in advancing the field of organized crime analytics, emphasizing the importance of such datasets in the absence of comprehensive real-world alternatives.