Auto-correlation and Fisher Face Algorithm based gender recognition using Voice and facial images

This work focuses on classifying speakers' gender based on their voice and facial features. These systems are useful for a number of things, including marketing, healthcare, user authentication, security, and more. We incorporated it to achieve high accuracy because gender recognition using voice or face, as implemented in the Uni-modal (facial images, voice samples) framework, suffers from poor accuracy. utilized a person's biological and behavioural traits as an input in this instance. The goal is to use the Auto-correlation method to estimate the speaker's gender from a speech sample, and Fisher face algorithm to identify the speaker's gender from a picture of their face, and integration of the results to improve the gender classifier's accuracy. A database containing voice and visual samples of both male and female pupils has been taken into consideration.