The national and international groups that address crime on a global scale say their new strategy to identify active criminals through nothing more than hearing their voice has passed its latest test, explains a new report in Joseph Farah’s G2 Bulletin.
While the idea may be scary to privacy advocates, the police forces of the world say they need a way to quickly identify a criminal suspect.
Their new effort is called the Speaker Identification Integrated Project, and authorities say it recently was tested by the United Kingdom’s Metropolitan Police Service and Portugal’s Policia Judiciaria.
More than 100 speaker-identification researchers and experts, forensic experts, intelligence analysts and police detectives, representing more than 50 nations, took part in the exercise.
Using a database of audio recordings, the demonstration showed how unknown speakers speaking in different languages can be identified through “social media or lawfully intercepted audios using a fusion of key markers such as gender, age, language and accent.”
The project has taken four years and has been funded by a grant of more than 10 million euros from the European Union.
It has involved 19 partners, including INTERPOL, Airbus, Singular Logic, Nuance, Saillabs Tech, Synthem and the University of Warwick and other police agencies and tech-industry leaders.
INTERPOL said that as a full project partner, it “focuses on ensuring the speaker identification technology meets the operational needs and requirements of law enforcement while guaranteeing that the legal aspects of the technology are compatible with existing national legislation including INTERPOL’s Rules for the Processing of Data and safeguards for individual privacy.”
The project has a website, siip.eu, which explains the variety of software features it uses.
They include “Speaker Model recognition,” gender detection, age detection, language and accent detention, and “Keyword & Taxonomy Spotting.”