August 10, 2010 at 14:56FSB learns to recognize Caucasians by voice
The FSB of Russia has announced a tender for "Creating an automated atlas of national languages of the Russian Federation" (R&D "D-2010-08-4.3") with an initial contract price of 24 million rubles. and for a period of 29 months ( tactical and technical task in PDF ). The order was placed on behalf of part 68240, whose affiliation with the FSB is known for analyzing information from open sources .
As part of the tender, it is required to develop a reference book, which can become the basis for an automatic system capable of reliably recognizing the language of the speaker by speech. For example, with the help of such a system, conversations in Caucasian languages can be quickly detected among all cellular conversations in Moscow (provided that they are simultaneously tapped through the switches of mobile operators).
The first languages for which "a study of the characteristics of the spoken language of informants-carriers" should be conducted, in the TTX six languages are called: Avar, Ingush, Kabardino-Circassian, Karachay, Balkar, Dargin. For each language, there must be at least 20 informants with different speech recording channels: microphone, telephone, etc., at least 10 recording sessions for each channel for more than 40 seconds. Then you need to analyze the sound recordings, and then compile linguistic passports of languages.
As part of the tender, it is required to develop a reference book, which can become the basis for an automatic system capable of reliably recognizing the language of the speaker by speech. For example, with the help of such a system, conversations in Caucasian languages can be quickly detected among all cellular conversations in Moscow (provided that they are simultaneously tapped through the switches of mobile operators).
The first languages for which "a study of the characteristics of the spoken language of informants-carriers" should be conducted, in the TTX six languages are called: Avar, Ingush, Kabardino-Circassian, Karachay, Balkar, Dargin. For each language, there must be at least 20 informants with different speech recording channels: microphone, telephone, etc., at least 10 recording sessions for each channel for more than 40 seconds. Then you need to analyze the sound recordings, and then compile linguistic passports of languages.