usvseg 1.0.2

Python ver of USVSEG

USVSEG for Python

  

This is a Python implementation of USVSEG. USVSEG is a GUI-based script for robust segmentation of rodents' ultrasonic vocalizations originally written in MATLAB.

See the repository of the original MATLAB version for general information about USVSEG.

Getting Started

To install the software in an Anaconda virtual environment, run the following command:

(base) C:\> conda create -n usvseg python=3.7 numpy scipy matplotlib tqdm pyyaml
(base) C:\> conda activate usvseg
(usvseg) C:\> pip install opencv-python soundfile
(usvseg) C:\> pip install usvseg

Alternatively, you can install the software in a Python venv virtual environment by running the following command:

C:\> python -m venv usvseg 
C:\> .\usvseg\scripts\activate.bat 
(usvseg) C:\> pip install numpy scipy matplotlib tqdm pyyaml PyQt5 opencv-python soundfile
(usvseg) C:\> pip install usvseg

To start the USVSEG GUI, run the following commands in the virtual environment:

(usvseg) C:\> usvseg

The GUI is compatible to the original USVSEG. You can find the usage of the GUI on the original site.

Contacts

Feel free to open an issue in github for questions and feature requests.

License

This software is released under the MIT License. See LICENSE for more information.

This python version of USVSEG is based on the original MATLAB version. See Third Party Notices for more information.

Citation

Original idea and processing procedures of USVSEG have been published in an open-access journal. Please cite this paper when you publish research works with using this software:

USVSEG: A robust method for segmentation of ultrasonic vocalizations in rodents.,
Tachibana RO, Kanno K, Okabe S, Kobayasi KI, Okanoya K
PloS one 15(2) e0228907