SReal time network, text, and speaker analytics for combating organized crime
Discovering criminal networks and identifying their members is one of the primary aspects of LEAs’ mission. ROXANNE contributes towards this goal by bridging the strengths of speech and language technologies (SLTs), visual analysis (VA) and network analysis (NA). ROXANNE will achieve a significant increase in the speed of investigation processes and an improvement in identification of individuals by means of speech, in the scope of criminal cases where large amounts of lawfully intercepted communications (with multilingual attributes) are analyzed.
The technical development will be centered around the ROXANNE platform, which will enhance criminal network analysis capabilities by providing a framework for extracting evidence and actionable intelligence based on speech, language and video technologies. The intention is not to replace humans but automate time-consuming tasks, and support LEA decision-making. Its early version will offer preliminary SLT, VA and NA capabilities to collect end-user feedback. The final version will provide multilingual, probabilistic tools interfacing SLT and NA technologies, boosted by natural language processing (NLP) and relation analysis in the synoptic criminal activity graph. ROXANNE will achieve full compliance with relevant INTERPOL and EU legal and ethical frameworks, including innovative approaches to data protection management such as privacy by design.
Special efforts will be expended to ensure ROXANNE outcomes achieve widespread adoption by law enforcement. The effort will be enhanced through a series of education and awareness campaigns and the direct involvement of LEAs from nine European countries, that will test our solutions on real case data. In addition, ROXANNE partner INTERPOL and EUROPOL (member of the External Advisory Board) will provide advice and guidance. The consortium has 24 partners with complementary skills, including leaders in key technology areas impacting criminal investigations.