Recognizing environmental sounds is a basic audio signal processing problem. Consider, for example, applications in robotic navigation, assistive robotics, and other mobile device-based services, where context aware processing is often desired. Human beings utilize both vision and hearing to navigate and respond to their surroundings, a capability still quite limited in machine processing. The first step toward achieving multi-modality is the ability to process unstructured audio and recognize audio scenes (or environments). The goal of this work is on the characterization of unstructured environmental sounds for understanding and predicting the context surrounding of an agent or device. This work investigates issues in characterizing unstructured environmental sounds and the development of appropriate feature extraction algorithm and learning techniques for modeling the variations of the acoustic environment.