PyRES

Python library for Reverberation Enhancement System development and simulation.

---

Installation

Follow the instructions below to install PyRES and to set up a working environment.

1. Clone the repository

git clone https://github.com/GianMarcoDeBortoli/PyRES.git
cd pyres

2. Install Python:

Make sure you have Python version >=3.10 installed on your system.

3. Set up the environment

• Automatic setup (recommended):
  • On MacOS/Linux (bash):

    bash bootstrap.sh

  • on Windows (cmd):

    call bootstrap.bat

• Manual Setup:
  • If you are using Conda:

    conda env create -f environment.yml --name pyres-env

  • If you are using Pip:
    • On MacOS (bash):

      brew install libsndfile
      python -m venv pyres-env
      source pyres-env/bin/activate
      echo "export DYLD_LIBRARY_PATH=$(brew --prefix libsndfile)/lib:$DYLD_LIBRARY_PATH" >> pyres-env/bin/activate
      python -m pip install --upgrade pip
      pip install -r requirements.txt

    • On Linux (bash):

      sudo apt-get update && sudo apt-get install -y libsndfile1
      python -m venv pyres-env
      source pyres-env/bin/activate
      python -m pip install --upgrade pip
      pip install -r requirements.txt

    • On Windows (cmd):

      python -m venv pyres-env
      .\pyres-env\Scripts\activate.bat
      python -m pip install --upgrade pip
      pip install -r requirements.txt

---

Tutorial

Please refer to the .examples/ folder for a series of tutorial files.

1. VrRoom class

   The VrRoom class represents the DSP architecture in a reverberation enhancement system.

   PyRES contains several architecture implementations.

   This tutorial shows the instantiation of a DSP and the most important attributes.

2. PhRoom class

   The PhRoom class represents the physical space in which the reverberation enhancement system is located.

   The physical space hosts the stage sources, the audience receivers, and the system transducers.

   PyRES, PhRoom has two subclasses:

   • PhRoom_wgn, which simulates the room impulse responses through exponentially-decaying white-Gaussian-noise sequences
   • PhRoom_dataset, which loads the room impulse responses from the accompanying dataset[1]

   These tutorials show the use and the most important attributes of the two.

3. RES class

   In PyRES the reverberation enhancement system is implemented by the RES class.

   The RES class receives an instance of VrRoom and PhRoom each, and controls the interaction between them.

   This tutorial shows the use and the most important attributes and methods of the RES class.

4. Training of a DSP

   PyRES relies on FLAMO[2] as backend for the signal processing.

   Thus, the DSP architectures are defined as chains of differential processing modules which can be trained through a machine-learning-like pipeline.

   These tutorials show the training of DSP architecture repeating previous work of the author[3][4].

---

References

[1] De Bortoli, G., Prawda, K., Coleman, P., and Schlecht, S. J. "DataRES: Dataset for research on Reverberation Enhancement Systems" (2.0.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15737243

[2] Dal Santo G., De Bortoli, G., Prawda, K., Schlecht, S. J., and Välimäki, V. "FLAMO: An Open-Source Library for Frequency-Domain Differentiable Audio Processing" Proceedings of the International Conference on Acoustics, Speech, and Signal Processing, pp.1--5, 2025

[3] De Bortoli, G., Dal Santo, G., Prawda, K., Lokki, T., Välimäki, V., and Schlecht, S. J. "Differentiable Active Acoustics: Optimizing Stability via Gradient Descent" Proceedings of the International Conference on Digital Audio Effects, pp. 254-261, 2024.

[4] De Bortoli, G., Prawda, K., and Schlecht, S. J. "Active Acoustics with a Phase Cancelling Modal Reverberator" Journal of the Audio Engineering Society, Vol. 72, No. 10, pp. 705-715, 2024.