In a bid to curb growing drug-related crimes, Mexican Interior Ministry plans to use one of the world's largest voice identification systems made by a Russian company to instantly identify voices in intercepted telephone conversations or any other audio messages.

The St Petersburg-based contractor company, Center of Verbal Technologies, has said "already now the system is capable of keeping in store and processing 600,000 voices." Officials said that by the end of the year the database was expected to exceed one million voices.

"Our system has become the world's first and, in fact, unparalleled voice biometric system of a national scale," claimed the director of the Center of Verbal Technologies, Mikhail Khitrov, adding that "proposals have been received for introducing similar complexes in other countries."

The system is capable of identifying the voice irrespective of the language. The identity will be established accurately no matter whether the suspect is ill, drunk or tries to change the voice intentionally, claimed officials

The Mexican police have already begun to collect a database of voice samples. The main equipment and software are at the data center of Mexico's federal police. The system's archive is available to more than 250 police offices across the country.

Voice identification systems began to be developed a long time ago. In Russia such experiments were made back in the 1950s at a classified research center of the NKVD security police.

The project's manager, Andrei Timofeyev, explained that "as it moves from the lungs to the lips, the air is filtered and distorted in a way that is very special for each human being. The tone is as unique as the fingerprints, which opens up vast opportunities for identification."

A large group of Russian mathematicians, phonetics specialists and programmers has worked for years to develop the system.