Synthesis of individualized HRTFs based on Neural Networks, Principal Component Analysis and anthropometry.
This repository was developed as part of a bachelor thesis project for Acoustic Engineering @ Federal University of Santa Maria - Brazil.
MATLAB app that can be used to generate individualized HRTFs with SOFA or HeSuVi extensions, using the proposed model.
MATLAB app that can create real-time virtual auditory scenes using SOFA HRTFs and n-channel audio input.
Contains the pre-processing and post-processing, regression model and HRTF reconstruction routines.
It may be necessary to ajust paths for your local directories.
Toolbox to work with SOFA HRTFs and general functions used by the main routines.