The aim of this thesis is to develop machine learning techniques that automatically classify audio signals into context classes, for outdoor and indoor sound, with the goal of recognising activity. In addition, techniques will be used to recognise and identify the speaker and their emotional state in voice signals.
