Academic Journals Database
Disseminating quality controlled scientific knowledge

Speaker Independent Recognition System with Mouse Movements

ADD TO MY LIST
 
Author(s): R.L.K.Venkateswarlu | R. Vasantha Kumari | A.K.V.Nagayya

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: ooc;
Issue: 1;
Date: 2012;
Original page

Keywords: Thread | Recognizer | Hidden Markov Model | Radial Basis Function | Mouse Movements

ABSTRACT
Speech recognition is potentially a multi-billion dollar industry in the near future. It is a natural alternative interface to computers for people with limited mobility in their arms and hands, sight, hearing limitation. For most current voice-mail systems, one has to follow series of touch-tone button presses to navigate through a hierarchical menu. Speech Recognition has the potential to cut through the menu hierarchy. Recently, neural networks have been considered for speech recognition tasks since in many cases they have shown comparable performance than the traditional approaches. There are two in-built threads in the recognition system. Thread 1 maintains the details about input acquisition where as thread 2 contains the classifier and decoder. The classifier used in this research is Radial Basis Function Neural Networks. The HMM graph is used as a decoder. The objective of the research is to make sure that the system is free from bugs. 100% accuracy is achieved by the recognition system.
Save time & money - Smart Internet Solutions     

Tango Rapperswil
Tango Rapperswil