The objective of this project is to develop a speaker recognition application using the Mel frequency cepstral coefficient (MFCC) and Gaussian mixture models (GMM). An application which will be able to perform biometric authentication of the users by using their voice as an input.
This speaker recognition application uses the Mel frequency cepstral coefficient for feature extraction. The features are extracted from a person’s voice with the help of Mel frequency cepstral coefficient (MFCC) technique. The application also uses Gaussian Mixture models and Expectation-maximization algorithm. Gaussian Mixture models (GMM) and Expectation-Maximization (EM) algorithm is used for the training of the application.