Date of Award

Fall 10-2016

Document Type

Thesis

Degree Name

Bachelor Degree

Department

Computer Science

First Advisor

Bertan Karahoda

Language

English

Abstract

Existing input methods for inserting text into digital devices are slow and have a steep contrast of information exchange difference between human and computer. While programmers quickly retrieve information from the computer visualy, via the screen, it is a slow process of inserting information via mouse and keyboard. Speech Recognition can be incorporated as another input method to enable inserting continuous text into the computer for increased speed. Accuracy improvements of speech recognition is driven by incorporation of deep neural networks and thus made speech recognition another viable input method, which is used in smartphones, computers and smart homes, to enable hand free technology usage by providing a voice user interface. This thesis presents HyperCode, a framework that makes use of Speech Recognition as another input method to increase coding efficiency, in the Java programming language, by using already present libraries that enable modification, extraction and alteration of the recognized words.

DOI

10.33107/ubt-etd.2016.1586

Download

Share

COinS