作者: John F. Sheets , Kim R. Wagner
DOI:
关键词: Communication device 、 Process (computing) 、 Speaker verification 、 Sample (statistics) 、 Authentication 、 String (computer science) 、 Voice Tag 、 Computer science 、 Speech recognition 、 Word (computer architecture)
摘要: Embodiments of the invention provide for speaker verification on a communication device without requiring user to go through formal registration process with issuer or network. Certain embodiments allow use captured voice sample attempting reproduce word string having random element authenticate user. Authentication is based both match score indicating how closely samples previously stored and pass fail response whether an accurate reproduction string. The processing network maintains history authenticated transactions samples.