Implementierung und Evaluierung von Modellen zur Schallquellenlokalisation

For sound localization the human auditory system uses binaural information like interaural time difference (ITD) and level difference (ILD) as well as spectral cues. The information can be described by individual head-related transfer-functions, HRTFs. They represent the spectral characteristics of the filtering process of an incoming sound wave by head, pinna and torso.

The processing of this information in the auditory system still cannot be fully explained. Hence the influence of several parameters on human perception is studied with the help of psychoacoustic experiments. However big variation in system parameters results in great expenses of time and money. Therefore auditory models which reproduce response patterns of human listeners are used.

The aim of the project is to combine two existing auditory models (Langendijk & Bronkhorst 2002 and Jepsen et al. 2008) and integrate the resulting one in the Auditory Modelling Toolbox (AMT) for MATLAB. The resulting model will be evaluated and validated with already existent data from several sound localization experiments.