This is an old version

Use the new version at https://github.com/theunissenlab/soundsep2

Sound Separation

Sound separation code for zebra finch vocalizations.

Initial setup

Creation of virtual environment for python dependencies

virtualenv env -p python3

Activation of virtual environment

source env/bin/activate

Install python dependencies

pip install -r requirements.txt

Create code -> src/main/python/ symlink

ln -s src/main/python code

On Windows

Installation of hdbscan requires Microsoft Visual C++ 14.0 or greater (get Microsoft C++ Build Tools from https://visualstudio.microsoft.com/visual-cpp-build-tools/). hdbscan might not be necessary though (used for clustering) so you could comment out the line in requirements.txt intsead.

From the terminal

python -m venv env
.\env\Scripts\activate
pip install -r requirements.txt
mklink /D code src\main\python

Extract potential calls from a file

This is a quick script that extracts calls from the first channel of a wav file. A more sophisticated extraction strategy from multichannel wav files or multiple simultaneously recorded wav files should use custom code based on this script.

From top level of the project (after installing dependencies and activating virtual environment), run

python scripts/process_wav_file.py <path_to_wav_file>

This creates a directory in the same folder as the wav file called "output" where the output numpy files will be saved.

Other setup options

Installation of QT Creator (only for GUI designer)

sudo apt-get install qttools5-dev-tools

Running the GUI

fbs run