C4.5: programs for machine learning
C4.5: programs for machine learning
| Hi-index | 0.00 |
Processing of acoustic signals with head-related transfer functions (HRTFs) is a commonly used technique for simulating 3D audio in interactive virtual environments and gaming applications. It is computationally very demanding to process the high dimensional HRTFs and yet capturing the variability across listeners and this could be a limiting factor for real time 3D audio applications. In this paper we show how the dimensionality of the HRTFs can be reduced to their minimum phase components, without losing the localization information contained in them, and their performance is evaluated under a localization classification paradigm using decision trees and compared with PCA.