Mesh2HRTF was developed at the Acoustics Research Institute as an open source project aiming at providing an easy-to-use software package for the numerical calculation of HRTFs. It targets researchers in the field of binaural audio, however the code for the boundary element method (BEM) can be used in numerous applications where the 3D Helmholtz equations needs to be solved numerically. In April 2023 the current Version 1.0.0 was released.

In a Nutshell

Mesh2HRTF simply reads geometrical data, calculates the corresponding sound field and produces HRTFs. To support multiple computer platforms, the concept of Mesh2HRTF is to focus on a command-line tool, which forms the numerical core, i.e., an implementation of the 3-dimensional Burton- Miller collocation BEM coupled with the multi-level fast multipole method (ML-FMM), and to provide add-ons for existing cross-platform applications for the preprocessing of geometrical data and for the visualization of results.

Features

  • Export of the 3D geometry of the head from blender ( https://www.blender.org/) to the inport format for the numerical part
  • Calculation of HRFTs using the boundary element method
  • Export of the HRTfs using SOFA

NumCalc

To calculate the HRTFs in Mesh2HRTF the BEM Code NumCalc is used. NumCalc is based on collocation with constant elements combined with a Burton Miller formulation to stabilize the calculations. Additionally, NumCalc uses the Fast Multipole Method to reduce memory consumption and computation time. As NumCalc uses a simple structured text file as input, it is possible, to use it not only for calculating HRTFs but also for more general problems involving the Helmholtz equation in 3D. An example input file NC.inp to calculate the scattering of a plane wave on a sphere can be found here: SphereExample.tgz. This tar-ball also contains an Octave script to generate the input file and an export skript for vtk to display the results of the calculations. The scripts use some commands like rows, and columns, which are currently not available in MATLAB. A version that also works with MATLAB will be available soon. With the release of Mesh2HRTF several features requested by users were included, e.g. an estimation of the memory requirements for a calculation.

Devolopment Team

Mesh2HRTF is currently developed and maintained in a collaboration of

Download

https://mesh2hrtf.org/