Code and Data


The software we produce lives in GitHub at

Hiearchical Grid Refinement (HiGRID): DOA Estimation using Rigid Spherical Microphone Arrays

higrid is a Python package that implements the hierarchical grid refinement (HiGRID) direction-of-arrival estimation algorithm based on steered response power density (SRPD) maps, and spatial entropy-based (multiple) peak detection. The code supplements our TASLP article Please consider citing that article if you find our code useful for your research.

The repository is self-contained and also includes the METU SPARG Eigenmike em32 Acoustic Impulse Response Dataset v0.1.0 (see below).

If you want to install the package using pip, execute pip3 install higrid in terminal.

The documentation of higrid is available here:


Gunshot Sound Recordings

Recordings of the sounds of four different pistols (Browning BDA 380, Glock 19c, Glock 21, Rossi Magnum 357) in an indoor firing range. The room impulse response of the firing range is also included. Data was used in:

Hacıhabiboğlu, H., (2017) “Procedural Synthesis of Gunshot Sounds based on Physically-motivated Models” in Game Dynamics: Best Practices in Procedural and Dynamic Game Content Generation, Oliver Korn, Newton Lee (Eds.), Springer-Verlag, 2017. external link

Please cite that paper alongside the dataset itself if you decide to use it.

METU SPARG Eigenmike em32 Acoustic Impulse Response Dataset v0.1.0

This dataset includes acoustic impulse response (AIR) measurements made using an Eigenmike em32 and the room impulse response measurements carried out at the same position using an Electron M6 measurement microphone. The measurements were made in classroom S05 at the METU Graduate School of Informatics on 23 January 2018. The full dataset along with its documentation is available on Zenodo as a 482 MB download.

Orhun Olgun, & Huseyin Hacihabiboglu. (2019). METU SPARG Eigenmike em32 Acoustic Impulse Response Dataset v0.1.0 (Version 0.1.0) [Data set]. Zenodo.