Evaluation toolbox for Sound Event Detection
Python Shell
Clone or download
toni-heittola Add evaluated_length_seconds as parameter for SegmentBasedMetrics's e…
…valuate method to allow use accuracy as a comparable metric (still not advisable metric for SED)
Latest commit e8b9690 Feb 2, 2018


sed_eval - Evaluation toolbox for Sound Event Detection

https://travis-ci.org/TUT-ARG/sed_eval.svg?branch=master https://coveralls.io/repos/github/TUT-ARG/sed_eval/badge.svg?branch=master

sed_eval is an open source Python toolbox which provides a standardized, and transparent way to evaluate sound event detection systems.



See http://tut-arg.github.io/sed_eval for manual and tutorials.

Installation instructions

The latest stable release is available on PyPI, and you can install with pip:

pip install sed_eval

Alternatively you can download or clone toolbox and use pip to handle dependencies:

unzip sed_eval.zip
pip install -e sed_eval


git clone https://github.com/TUT-ARG/sed_eval.git
pip install -e sed_eval


The toolbox is tested to work with Python 2.7 and Python 3.6.

  • numpy >= 1.7.0
  • dcase_util >= 0.1.6


If you use sed_eval in a research project, please cite the following paper:

Annamaria Mesaros, Toni Heittola, and Tuomas Virtanen, "Metrics for polyphonic sound event detection", Applied Sciences, 6(6):162, 2016 [HTML][PDF]


Code released under the MIT license.