Biometric Authentication: A Machine Learning Approach

10.5. Audio and Visual Biometric Authentication

Audio and visual (AV) biometrics has long been an active area of research, primarily because of the promise it can bring to practical applications. For example, voice biometrics can suffer severe performance degradation under a noisy acoustic environment, but facial images are unaffected. Conversely, facial image quality can be severely affected in poor lighting conditions, but lighting has no effect on voice quality. These two independent and complementary information sources are ideal candidates for enhancing biometric system reliability. With the recent introduction of third-generation mobile services, the use of both voice and facial images for biometric authentication has become practical and ...