Skip navigation
st. Mary's University Institutional Repository St. Mary's University Institutional Repository

Please use this identifier to cite or link to this item:
Title: Blind Assistant System (Lamba)
Authors: Eyuael Bezabeh, Helen Girma, Tomas Beyene, Zerihun H/Michael
Issue Date: 1-Sep-2022
Abstract: Vision is a fundamental part of our lives. Information is present everywhere in different forms, but most of it is processed by the visual organ, the eyes. In this sense, blind people have great difficulty for getting information from the environment. Especially in today’s technologically developed world, how they can cope up with is a great challenge. Locally, almost no product is found that aids blind people in navigating and interacting with technological tools. The main objective of this paper is designing an efficient blind assistance system for the visually impaired people using Artificial Neural Network (ANN) with modern cloud service. ANN is based on a supervised procedure and comprises three layers: input, hidden, and output. The system logically has two phases. The first is Automatic Speech Recognition (ASR). This class of application starts with a clip of spoken command audio in Amharic language and extracts the words that were spoken, as text. Not only do they extract the text but they also interpret and understand the semantic meaning of what was spoken, so that they can respond with answers, or take actions based on the user's commands. This phase is based on Artificial Neural Network (ANN) Model that trains with audio Dataset in order to identify the input audio commands. The second phase, Speech synthesis is where artificially produced human speech delivers the information requested by the blind person. Through the system, blind individuals will peruse basic technological services.
URI: .
Appears in Collections:the 16th National Student Research Forum

Files in This Item:
File Description SizeFormat 
Eyuael Bezabeh,Helen Girma,Tomas Beyene,Zerihun Hailemichael.pdf1.11 MBAdobe PDFView/Open
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.