Audio signal analysis for classification and source localization in e-learning applications.

Abstract

E-learning provides an opportunity for the students in a classroom to receive and participate in lectures broadcasted from a remote location. For convenience and scalability it is preferable to have a microphone array instead of an individual microphone per user to capture the signals. A Direction of Arrival (DOA) system can be used to determine the angle of arrival of the speech signal so that a camera can be steered in the direction of the signal source. In a DOA system for e-learning, Time Delay Estimation (TDE) techniques are suitable to determine the relative time difference between the signals to reach the microphones in the array. This time difference can be used to find the Direction of Arrival (DOA) of the signal. This thesis presents a DOA system called Modified yIN based Doa Estimator for e-leaRning (MINDER).