Skip to content

Latest commit

 

History

History
30 lines (22 loc) · 1.28 KB

README.md

File metadata and controls

30 lines (22 loc) · 1.28 KB

Geometric and Visual Artificial Reverberation

This project uses geometry mapping and image recognition techniques to create artificial reverberation. The implementation leverages the MATLAB fdnToolbox and Python.

Prerequisites

MATLAB Dependencies

  • MATLAB: Ensure MATLAB is installed on your system.
  • fdnToolbox: GitHub.
    • No need to download as the code is included in this repo.
  • MATLAB vs-code extension: Use the VS Code extension to add the required files to the MATLAB folder. To do this, right-click on the folder and select MATLAB: Add Folder To Path.

Python Dependencies

Install the following Python packages using pip3:

pip3 install numpy scipy pyroomacoustics matlabengine matplotlib librosa sympy openai

Documentation

  • Add open API key to utils/api_key.py file.
  • Configure config.py with paths for surface images and .obj format 3D model.
  • Test geometry recognition via: python3 model.py
  • Test material extraction via: python3 absorption.py
  • Create RIRs with all models via: python3 rirs.py
  • Create processed samples with all models via: python3 stimuli.py
  • Run Parallel ISMFDN via: python3 parallel_ism_fdn.py
  • Run Serial ISMFDN via: python3 serial_ism_fdn.py