Wednesday, April 22, 15:00 - 17:30, Location: Show and Tell Area A
Markus Kallinger, Giovanni Del Galdo, Fabian Kuech, Dirk Mahne, Richard Schultz-Amling
The proposed demonstration shows a novel spatial filtering procedure, which is integrated into the well-known Directional Audio Coding (DirAC) spatial codec. Spatial filtering works in the coded domain of DirAC and is performed by a short-time spectral attenuation technique, whose weights are calculated by means of the DirAC parameters, i.e., direction-of-arrival and diffuseness of the sound field. One major benefit of embedding spatial filtering into DirAC is the capability of rendering its output to almost any loudspeaker or headphone setup.
The demonstration consists of a small microphone array, a soundcard with microphone-preamps, a notebook and closed headphones. The PC-based realtime software performs the DirAC encoding, spatial filtering and rendering for headphone playback. The GUI allows - along with numerous other options - for rotating the desired look-direction and changing the beamwidth of the spatial filtering processing block. Headphone playback reveals one of the major benefits of the overall structure: residual portions of the suppressed spatial sound sources remain at their original positions - the intelligibility of a desired speaker is enhanced by two properties of the system: spatial suppression of disturbing sources and spatially correct reproduction of a sound field.