Files in this item



application/pdfPooja_Jain.pdf (494kB)
(no description provided)PDF


Title:Visualizing speech production with a hidden Markov model tracker to aid speech therapy and communication
Author(s):Jain, Pooja
Advisor(s):Karahalios, Karrie G.
Department / Program:Computer Science
Discipline:Computer Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):speech delay
communication disorder
hidden Markov model
Abstract:Communication disorders occur across all age groups of people and often show first signs of appearing in children. These can range from problems in comprehension of speech to expression of speech to the point that it interferes with an individuals achievement and/or quality of life. Communication disorders can compromise a persons psychological, sociological, educational and vocational growth. There have been various studies on how the implications of these impairments can be mitigated through treatment, therapy and communication processes. This research focuses on the development and implementation of a software that aims to facilitate speech production by providing feedback through audio visualizations that represent basic audio features and coherent parts of speech tracked by a hidden Markov model. The goal of these visualizations is to help the user understand speech better by providing a system where users can see the words they speak and experience, develop and practice speech skills using the statistical speech model and temporal features represented through simple abstract visualizations. This research proposes an approach to visualize speech in a way that can potentially aid speech therapy and communication to help people with communication disorders by providing them with a tool they can use to understand their speech problems without the continuous need of a therapist or teacher.
Issue Date:2013-08-22
Rights Information:Copyright 2013 Pooja Jain
Date Available in IDEALS:2013-08-22
Date Deposited:2013-08

This item appears in the following Collection(s)

Item Statistics