Welcome to the upgraded BRAC University Institutional Repository. We are currently organizing collections after a recent system upgrade. Homepage category counters may temporarily show lower numbers while syncing, but over 27,000 repository items remain safe and accessible. Please use the search bar to find theses, scholarly outputs, and institutional documents.

Implementation of Bangla speech recognition system on cell phones

Citation

Abstract

Speech Recognition refers to the process of converting analogue speech signals into text. Since the 1970s a lot of work has undergone in this particular field. The complex nature of speech due to it's contextual meaning , dialects, accents as well as the environment makes the task of recognizing speech verydifficult. A lot of research is currently under progress to increase the accuracy of this system. Although extensive research has been done on other language, the field of automated speech recognition in Bangla is at it's early level. With almost 200 million Bangla speakers all over the world, there can be a lot of applications of speech recognition. Cellular phone is one such domain , where voice commands can be used to initiate tasks. High accuracy can be achieved as the number of commands are rather limited and since it is dependent upon the user's voice, extensive training will yield better results. If an application is developed for cellular phones that can execute Bangla voice commands, it will make it easier for a large number of disabled people to use cell phones in their own languages. Since a lot of the users are not literate about how to use commands by pressing keys of phones will be facilitated by this application. However, the limited processing power of the cell phones are also a challenging factor. The aim of this thesis was to show a way to develop an application that will be able to recognize Bangla voice commands for cell phone and execute them followed by an attempt to implement the system according to research findings and analyze it.

Description

Cataloged from PDF version of thesis report.
Includes bibliographical references (page 43).
This thesis report is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2011.

Publisher Link

Type

Thesis