. "RIV/49777513:23520/04:00000091" . "Comparison of several speaker verification procedures based on GMM" . "4"^^ . "RIV/49777513:23520/04:00000091!RIV07-GA0-23520___" . "1777" . "Padrta, Ale\u0161" . "speaker verification, recognition algorithms"@en . . . "Porovn\u00E1n\u00ED n\u011Bkolika procedur verifikace \u0159e\u010Dn\u00EDka zalo\u017Een\u00FDch na GMM"@cs . . . . "Comparison of several speaker verification procedures based on GMM"@en . . "[223C7A0F60BA]" . "KR - Korejsk\u00E1 republika" . "2004" . . "2"^^ . "558290" . "0" . "23520" . . "Porovn\u00E1n\u00ED n\u011Bkolika procedur verifikace \u0159e\u010Dn\u00EDka zalo\u017Een\u00FDch na GMM"@cs . . "In this paper, three speaker verification procedures are tested. All the procedures are based on Gaussian mixture models (GMM), however, they differ in the way, in which they use particular feature vectors of an utterance for speaker verification. A lot of experiments have been performed in a group of 329 speakers. The results showed that there is a procedure that enables to achieve better results than the commonly used procedure based on the log likelihood of the whole utterance - the procedure based on the majority voting rule for single feature vectors." . "Comparison of several speaker verification procedures based on GMM"@en . "2"^^ . "1225-441X" . . "Journal of the Acoustical Society of Korea" . "Radov\u00E1, Vlasta" . . . . "Comparison of several speaker verification procedures based on GMM" . "P(GA102/02/0124), Z(MSM 235200004)" . . . . "V \u010Dl\u00E1nku jsou testov\u00E1ny 3 procedury verifikace \u0159e\u010Dn\u00EDka. V\u0161echny procedury jsou zalo\u017Eeny na Gausovsk\u00FDch hustotn\u00EDch sm\u011Bs\u00EDch (GMM), li\u0161\u00ED se ale ve zp\u016Fsobu, jak\u00FDm pro verifikaci vyu\u017E\u00EDvaj\u00ED jednotliv\u00E9 p\u0159\u00EDznakov\u00E9 vektory. Procedury byly testov\u00E1ny na mno\u017Ein\u011B 329 \u0159e\u010Dn\u00EDk\u016F. V\u00FDsledky uk\u00E1zaly, \u017Ee metoda zalo\u017Een\u00E1 na v\u011Bt\u0161inov\u00E9m hlasovac\u00EDm pravidle pro jednotliv\u00E9 p\u0159\u00EDznakov\u00E9 vektory umo\u017E\u0148uje dos\u00E1hnout lep\u0161\u00EDch v\u00FDsledk\u016F ne\u017E be\u017En\u011B u\u017E\u00EDvan\u00E9 metody zalo\u017Een\u00E9 na logaritmu prav\u011Bpodobnosti cel\u00E9 promluvy."@cs . "In this paper, three speaker verification procedures are tested. All the procedures are based on Gaussian mixture models (GMM), however, they differ in the way, in which they use particular feature vectors of an utterance for speaker verification. A lot of experiments have been performed in a group of 329 speakers. The results showed that there is a procedure that enables to achieve better results than the commonly used procedure based on the log likelihood of the whole utterance - the procedure based on the majority voting rule for single feature vectors."@en .