Capturing blocked-entrance binaural signals from open-entrance recordings

Binaural recordings enable us to capture all sound attributes such a source characteristics, environment effect and spatial information of a given acoustic environment. It has been shown that blocked-entrance binaural recordings provide advantages over open-entrance recordings, primarily because blocked-entrance recordings are not influenced by the ear-canal acoustics of the individual for which it is recorded. However, blocking the ear canal impairs normal hearing conditions, which may be unacceptable for applications in which binaural audio capturing is desired, but, without interfering the individual's hearing and doing. This project is concerned with the development of a strategy for binaural audio capturing as open-entrance recordings (minimum interference), and for transforming these recordings to blocked-entrance versions that are more suitable for analysis and reproduction of binaural audio in a more general context. The first point involves the development of a wearable binaural recorder. The second point involves implementation of transfer-function measuring techniques, e.g. maximum-length sequence (MLS) with focus on in-situ measurements, and the use of inverse filtering techniques to compute filters that will allow transforming open-entrance recordings to blocked-entrance recordings. This work is supported by the SKILLS integrated project.