Assamese Speech Recognition using hybrid HMM and ANN

This book deals with the development of phone based Assamese speech recognition using hybrid HMM and ANN. Here a baseline system using HMM have also been discussed. The baseline system served as a bench mark for the main hybrid HMM and ANN system. Based on the results obtained it is concluded that the hybrid system give a better recognition rate as compared to the conventional HMM based system. The limitations of HMM and ANN which results in the hybrid system is also mentioned. For development of the both the systems the mel cepstral feature extraction is used as the feature extraction technique. The details steps involve in its extraction is also discussed.

Autorentext

Geetima Kachari received her B.E and M.Tech degree from RGPV,Bhopal and NEHU,Shillong in Electronics and Communication Engineering in the year 2009 and 2015 respectively. She is currently working as Laboratory Engineer in Speech and image processing lab, Department of Electronics and Communication Engineering,NEHU, Shillong.