Software Repository - Spatial Audio Scene Characterization (SASC): Automatic Localization of Front, Back, Up, and Down-Positioned Music Ensembles in Binaural Recordings
This repository consists of scripts and data useful to replicate the experiments described in the paper "Spatial Audio Scene Characterization (SASC): Automatic Localization of Front, Back, Up, and Down-Positioned Music En-sembles in Binaural Recordings".
The repository is organized as follows:
- models - the deep learning models that were trained in the study
- scripts - scripts used in the development of the deep learning algorithm: finding hyperparameters, model traning, evaluation, visualization, and statistical calculations
- data - partial results and detailed description of the data used in the study
Software dependencies:
- MATLAB - a development environment used to implement models
- VOICEBOX - a toolbox used to implement the binaural convolver
- SOFA - a file format for reading, saving, and describing spatially oriented data of acoustic systems.
Sławomir K. Zieliński 1, Paweł Antoniuk 1, and Hyunkook Lee 2
1 Faculty of Computer Science, Białystok University of Technology, 15-351 Białystok, Poland; s.zielinski@pb.edu.pl (S.K.Z.); p.antoniuk6@student.pb.edu.pl (P.A.)
2 Applied Psychoacoustics Laboratory (APL), University of Huddersfield, Huddersfield HD1 3DH, UK; H.Lee@hud.ac.uk (H.L.); D.S.Johnson2@hud.ac.uk (D.J.)
The content of this repository is licensed under the terms of the GNU General Public License v3.0 license. Please see the LICENSE file for more details.