"CASA is designed to train researchers in the development of algorithms, codecs and tools for the 3D audio technology that will be required to deliver very high-quality, object driven sound for immersive and interactive entertainment. Audio technologies and coding methods are as essential to the development of very high quality immersive and interactive 3D media as MP3 (MPEG-1 Audio Layer 3) has been to on-line music and streaming media. Sound is directional: the positioning of sound sources and the acoustics of the space are critical to the experience, impact and sense of presence. Future 3D media and the Internet will need to make it possible to locate and navigate 3D sound as well as vision. Current channel-based approaches to audio coding are unsuitable in several respects and object-based schema such as MPEG-4 AudioBIFS (ABIFS) and Advanced AudioBIFFS (AABIFFS) have so far proved too complex to implement. CASA will research an develop a new 3D audio object approach, based on the application of Auditory Scene Analysis and metadata-based processing to create algorithms, codecs and methods for delivering multi-channel audio that can support 3D interaction. The resulting tools will make it possible for audio engineers to capture and transmit the audio mix as metadata, which can be transmitted with audio assets in such a way as to recreate the audio scene in an optimum way, in any digital environment be it a smart phone, PCs, games console, in-car or home entertainment system, or cinema. The project will be implemented in a programme of work built around a series of exchanges of scientific and technical staff between an academic research centre Dublin Institute of Technology (DIT, Ireland), specialising in audio research and algorithm development; Europe’s leading audio codec developers at DLLNI Ltd (UK) and a specialist SME audio postproduction company (Tamborine Ltd, UK)."