This project aims to develop an AI-powered hearing aid system
designed to enhance speech comprehension in noisy environments
for individuals with hearing loss. By employing machine learning
algorithms, the system will effectively suppress background
noise while amplifying speech signals, enabling clearer
communication. Through extensive data collection and deep
learning techniques, the system will be trained to generalize
across various real-world acoustic situations for optimal
performance.
This theme will also create a cutting-edge artificial intelligence
(AI) hearing aid system that is especially designed to improve
speech understanding in noisy settings for people who have hearing
loss. The project aims to use machine learning algorithms to
accurately identify and suppress noise while preserving and
amplifying speech signals, thereby enabling clearer and more
understandable communication for users in difficult acoustic
environments. This will be achieved by leveraging advances in deep
learning techniques. Complex machine learning models, such as
classification algorithms, which have been trained on a range of
datasets with different speech and noise characteristics. To
enable these models to generalize well across various contexts and
noise situations, a rich dataset that captures the intricacy of
real-world auditory sceneries through substantial data collection
and annotation activities will be built.
PROJECT TEAM
Dr. (Mrs.) Rose-Mary Owusuaa Mensah Gyening
Lecturer, Department of Computer Science, Kwame Nkrumah University of Science and Technology (KNUST) in Ghana