Signal processing techniques used in speech recognition or synthesis. An overview

The paper presents an overview of the techniques used in audio stream processing in order to obtain a sufficiently complete knowledge base to be used in speech recognition or synthesis. At the beginning, some of the most common methods are presented in tabular form. The mathematical representation of Mel Frequency Cepstral Coefficient (MFCC), Gaussian Mixture Model (GMM), Hidden Markov Model (HMM), and K-Nearest Neighbor Classifier (KNN) is reviewed.