Strategies for template-free direct biometric encryption using voice based features

Current biometric systems references information stored on templates and it possess major drawbacks in their inability, if lost or stolen, to be revoked and re-issued as would be the case with passwords. Thus, once a biometric source has been compromised, the owner of the biometric, as well as the data protected by the biometric, is compromised for life. This research investigates the potential of the voice modality for employment in a template-!ree biometric system. which requires storage of neither hiometric templates nor encryption keys. It 'rl'orks by directly encrypting the biometric data provided by the human voice and therefore eliminates the need for storing templates used for data validation, and thus increasing the security of the system. The research also introduces feature concatenation as a novel method of combining the binary information from the feature sets and also introduces five new revocation strategies based on the template-free method The promising results allow us to conclude that voice may potentially form the basis for a practical template-free biometric system.