Local Features (LFs) Based Bangla Phoneme Recognition

This monograph discusses the dominance of Local Features (LFs), as input to the Multilayer Neural Network (MLN), extracted from a Bangla input speech over Mel Frequency Cepstral Coefficients (MFCCs). Here, LF-based method comprises three stages- (i) LF extraction from input speech, (ii) Phoneme probabilities extraction using MLN from LF and (iii) The Hidden Markov Model (HMM) based classifier to obtain more accurate phoneme strings. In the experiments on Bangla speech corpus prepared by us, it is observed that the LF-based Automatic Speech Recognition system provides higher phoneme correct rate than the MFCC-based system. Moreover, the proposed system requires fewer mixture components in the HMMs. Moreover, this paper reviews some of the key advances in several areas of automatic speech recognition. We also illustrate, by examples, how these key advances can be used for continuous speech recognition of Bangla. Finally we elaborate the requirements in designing successful real-world applications and address technical challenges that need to be harnessed in order to reach the ultimate goal of providing an easy-to-use, natural, and flexible voice interface between people and machines.

Autorentext

Mohammad Nasiruddin has obtained his Masters degree in Computer Science & Engineering in 2011. Since then he has worked in different international projects in the field of Bangla Speech Recognition & Software Engineering. He is the author of several articles published in reputed journals and is a member of different in international working groups.