egrinstein / xsrp

Modular implementation of the Steered Response Power method and its variants

XSRP: eXtensible Steered Response Power

This repository contains the code for the paper:

Steered Response Power for Sound Source Localization: A Tutorial Review

The following functionality is currently implemented:

• Conventional SRP-PHAT in the time domain (taking the DFT, applying the phase transform, followed by the IDFT) as described by Dibiase et al. [1]
• Conventional SRP-PHAT in the frequency domain (same as above without the IDFT step) as described by Dibiase et al.
• SRP in the time domain, using temporal cross-correlation without phase transform
• Parabolic interpolation of the cross-correlation function in time
• A simple Volumetric SRP approach which projects the average of N-closest correlation values instead of only the one associated with the microphone pair's Time Difference of Arrival (TDOA)
• Grid creation functions for Positional Source Localization and Direction of Arrival (DOA) Estimation
• Visualization tools

Installation

We recommend installing a Conda virtual environment using the provided environment.yml file. This will install all the required dependencies. Change directory to the xsrp folder and run the following commands:

1. conda env create -f environment.yml
2. conda activate xsrp

You can then optionally run the tests to verify that everything is working correctly:

3. python -m pytest tests

You may want to check the images that were generated in the tests/temp folder.