Models and algorithms for binaural hearing (B04)

点击翻译按钮获取中文摘要

A functionally simple and computationally fast model of human binaural processing and spatial audio processing will be developed. The binaural model is going to be in functionally equivalent to human auditory processing but without modelling action potentials or even the underlying biophysical processes. The model should be able to explain a broad range of psychoacoustic data. In contrast to delay-line models, it does not need the computation of the running cross-correlation, allowing for real-time processing. A key extension to be developed within this project is a localization algorithm mimicking the spatial impression a normal hearing listener has to any given sound input.