The first Roxanne field test took place

Roxanne is a EU-funded project that leverages text, speech and video in real-time in order to build tools for combating organized crime.

In this video, we introduce the main technologies behind our platform, and their role in investigations conducted by Law Enforcement Agencies (LEAs). Whenever a phone call is intercepted, the language of the call is identified, and Automatic Speech Recognition (ASR) provides transcripts of the call. Based on these transcripts, we analyze the content of the conversation, the topics, the places, the dates mentioned... We also assess the identity of the speakers involved in the phone call using Open-set speaker identification (SID) systems. Once we predicted the identities of the speakers and the content of the call, we store these information in a criminal network. The network represents the links between the various speakers, and allows investigators to understand links between suspects and groups. Finally, using Social Network Analysis (SNA) tools, we highlight the most central characters of a case or the various sub-groups in the network. The first field test of ROXANNE was a great success, with over 80 people joining from various LEAs.

Prepared video summarizes the field-test demonstration can be seen here.


More information

- Roxanne project website