SPiD system recognizes individuals from their voices whereas the LiD engine identifies the language spoken by a particular person. Both systems have successfully been deployed in surveillance and security screening applications. These systems are equally suitable for use in remote authentication systems hosted by payment operators and online service providers. Voice recognition provides a convenient and low-cost alternative to token and SMS based second-factor authentication.
SPiD and LiD comprise of state-of-the-art sub-systems for performing audio preprocessing, spectral analysis, feature extraction, voice parameterization, discriminative classification, high-speed kernel computation and system administration.
