Description:
Recently there has been a significant amount of advancement in the area of speech recognition. Technology now exists that will transcribe words onto a computer screen with accuracy rates of 90% or better.

Here at Villanova, Dr. Tom Way and his associates have been working on an application that uses this technology in the classroom. Our goal is to produce a system that can transcribe a professor's lecture onto his student's laptops in real time. Currently the system that is in place is built on Java using a client/server architecture.

My project aims to do a couple of things. First, I will redesign the server piece of the application using the .NET framework. This will allow me to tap into the new microsoft speech API that was recently released. Also, by using my personal computer for development, I will be able to take advantage of the Windows Vista speech recognition engine. Along with these improvements I hope to develop a more advanced transcription algorithm than the system uses in its current form. My goal is to produce better accuracy with these enhancements.

Project Schedule

Date	Description	Status
December 2007	Meet with advisor, finalize topic, register for independent study	Complete
1/14	Meet with advisor, flesh out details of work	Complete
1/21	Create Project Website	Complete
1/27	Successfully Build and Run project in java/netbeans	Complete
1/28	Enable Speech Recognition on PC and go through standard training	Complete
2/2	Complete prototype program of speech recognition in C#	Complete
2/13	Meet with professor	Complete
2/24	Complete migration of existing server code from Java to C#	Complete
3/12	Add new enhancements to VUST code	Complete
3/19	Meet with professor	Complete
3/24	Begin testing - Create sample audio files	Complete
4/2	Complete all testing	Complete
4/16	Paper Rough Draft	Complete
5/2	Paper Final Draft	Complete