Author : Akshatha P 1
Date of Publication :7th June 2017
Abstract: Speech recognition is one among the most recently developing field of research, at both industrial and scientific levels. In this paper we have developed a gadget assistance system for an old age home, where the operational commands for a TV was provided as inputs and tested for its performance. This system is configured for both online and offline speech recognition. The online speech recognition uses DNN algorithm, whereas the offline speech recognition uses HMM algorithm. This gadget is also used for automation of devices such as television with the help of a LIRC module. Various speech recognition modules for playing Music, Videos, Games and Television control were written and executed. Accuracy of speech recognition was observed.
Reference :
-
- D. Yu and L. Deng”Automated Speech Recognition: A Deep Learning Approach, “ Springer Pub., 2014
- Claudio Becchetti and KlucioPrinaRicotti. “Speech Recognition : Theory and C++ implementation," Wiley-Blackwell pub., 1999
- Douglas A Renoylds and Richard C Rose, ”Robust text- independent speaker identification using gaussian mixture speaker models," IEEE Transactions on Speech and Audio processing, Vol. 3, No.1, pp72-83, 1995