The first, SAFLink's SAFtyLatch uses the US government-developed Human Authentication application programming interface (HA-API) to enable it to work with more than one biometrics device. The second, Net Nanny's BioPassword LogOn for Windows NT uses an unusual biometric -- typing rhythm -- to increase Windows NT security. SAFLink's SAFtyLatch, which will ship by the end of the year in the US for $59.95 and shortly afterward in the UK, uses HA-API-compliant voice recognition to encrypt or decrypt files and folders on Windows 95 or Windows 98. HA-API, which was developed by the US Department of Defence this year, abstracts the differences between biometrics devices, enabling applications to use many different devices with a single API. Although the first version of SAFLink uses voice as an authentication method (using an HA-API-compliant engine by Lernout & Hauspie), HA-API will enable the product to work with many of other devices, according to SAFLink's chairman, Jeffrey Anthony. SAFLink also demonstrated its Multi-Biometric Product Suite, which lets companies or developers use many biometrics devices with Windows NT, Web sites, CA's Unicenter and Novell's NetWare. Each product in the suit already supports many different devices, including six fingerprint readers and two facial recognition systems, according to Anthony. Net Nanny's BioPassword, on the hand, enables corporations to deploy biometrics technology without any biometrics devices at all. It uses the pattern of typing, which varies just enough between individuals to be used as a biometric. BioPassword is a new subsidiary of Net Nanny. The BioPassword LogOn for Windows NT product, which will be available in the first quarter of next year, users will still be required to use passwords to log into Windows NT. BioPassword, however, replaces NT's normal login module with one that also watches typing speed and rhythm to determine if the right person is logging onto the network. According to Tom Yerex of research and development, the idea of using typing patterns as a biometric came from research done several years ago at Stanford University. It gives companies an extra level of product against bad, easily-guessed passwords, Yerex said. "We believe the technology can be used with almost any device," Yerex said. In the future, the company may adopt the method for use on the number pads at cash points, for example. A demo of the technology is available on the web at www.biopassword.com