Scroll to Content
Group Members
Aaron Andrews, Harvey Barber, James Bradley-Webster, Thomas Dent, Peter Higgins
Supervisors
Dr Keith Holland, Dr Takashi Takeuchi
Supporters
OPSODIS Limited
Many modern entertainment systems use spatial audio to enhance the user experience during activities such as watching films or playing video games. Conventionally, spatial audio is delivered through numerous loudspeakers positioned around the room. Loudspeaker arrays can allow accurate spatial audio to be reproduced without the need for large numbers of loudspeaker units.

Two common spatial audio techniques are cross-talk cancellation (CTC) and beamforming. CTC is a signal processing method using loudspeakers to isolate left and right channel audio signals at the respective ears. The result is like listening through headphones, but with improved comfort and the capability of entertaining multiple listeners. Beamforming is a technique where sound is steered as a beam towards a desired angle, giving an acoustic bright zone, with an acoustic dark zone outside the beam. Projects featuring these techniques have often utilised arrays with equal spacing between each driver. Research papers suggest that an array with driver spacings which increase logarithmically, may present benefits to CTC efficiency, as well as giving more consistent beamformer properties at different frequencies.

This project explored these benefits by designing, constructing and testing a logarithmically spaced array to be used for CTC and beamforming. Initial estimations of key parameters were obtained through computational models, before an iterative design process was used to optimise the arrays performance. Finally, the suitability of the design as an entertainment system was tested via a series of objective and subjective tests. Participants gave responses characterising the accuracy of spatial audio reproduction, providing valuable insight into the performance of the design.