The book in hand provides the reader with the basics of speech as well as in-depth technicalities of accurate recognition by use of two different artificial intelligence techniques. The literature review presented in a number of chapters, refines user's concepts about the phonetics and mel frequency ceptstral cofficients that characterize speech as well as voice. These coefficients are then operated upon in MATLAB for analysis and classification to one of the predefined classes of a trained system. After reading this book thoroughly the user will be able to conclude that the efficiency of recognition is enhanced when two artificial intelligence or classification techniques, called Linear Discriminant Analysis and Support vector machines, concatenate. This book also gives brief knowledge about the Mercer Kernel, k-nearest neighbor and many other technical concepts.

This book is the result of a huge effort by the author to complete this project for submission as a dissertation for the degree of M.Sc in Electronic Communication and Computer Engg at The University of Nottingham.Currently, the author is working at a public sector university in Pakistan with research interests in FPGA's and speech recognition.