Back
|
Print

ICT Law Projects

Speaker Identification Integrated Project (SIIP)

Siip

INTERPOL is participating in a research project to develop new technology to help the law enforcement community identify the voices of unknown individuals.

The project is conducted by an international consortium of 19 partners comprising end-users, industry and academia.

A new forensic capability to help solve crime

Speaker identification technology can help rapidly identify suspects’ voices and isolate conversations of interest in a wide range of law enforcement cases, such as match-fixing, ransom or terrorist calls.

The SIIP project will help identify criminals and terrorists through the development of a speaker identification engine able to run on different sources including the Internet, via social media, or fixed and mobile telephones in lawful interceptions.

In addition to voice recognition, the SIIP system will help identify gender, age, language and accent, and detect voice cloning.

In the future, speaker identification could be combined with other biometric technologies such as fingerprint and facial recognition and thereby facilitate checks against all related databases.

Data protection

The SIIP project explores the operational, technical, forensic, legal and ethical perspectives related to the evolving field of speaker identification.

INTERPOL’s role is to ensure the technology meets the operational needs and requirements defined by the law enforcement community in its member countries.

This will focus in particular on the legal aspects of the technology, to ensure it is compatible with existing national legislations and in line with current privacy and data protection laws, including INTERPOL’s Rules on the Processing of Data.

Funding

This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement No. 607784.

Launched in 2014, the project will run for four years.

Speaker Identification Integrated Project video

Event