Improved Binaural Model for Localization of Multiple Sources

Konferenz: Sprachkommunikation - Beiträge zur 10. ITG-Fachtagung
26.09.2012 - 28.09.2012 in Braunschweig, Deutschland

Tagungsband: Sprachkommunikation

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Schäfer, Magnus; Bahram, Mohammad; Vary, Peter (Institute of Communication Systems and Data Processing, RWTH Aachen University, Germany)

An improved binaural hearing model is proposed that consists of a physiologically motivated signal processing step and a subsequent cognitive model. The model is shown to be capable of correctly detecting and tracking sources while blindly determining the number of active sources based on temporal and spectral information. The complete model is of interest in all areas that need to consider the capabilities of the human hearing system while the ability to determine the number of active sources makes it a logical enhancement for source separation and spatial signal processing such as adaptive beamforming.