Detection of overlapping acoustic events using a temporally-constrained probabilistic model

Overview

Description: This code performs sound event detection in complex acoustic environments. This version is trained with office sounds (16 event classes) from the DCASE 2013 Office Synthetic (OS) challenge: http://c4dm.eecs.qmul.ac.uk/sceneseventschallenge/ . The output is a list of detected events per onset, offset, and event class.

If you use this code, please cite the following paper:
E. Benetos, G. Lafay, M. Lagrange, and M. D. Plumbley, "Detection of overlapping acoustic events using a temporally-constrained probabilistic model", IEEE International Conference on Acoustics, Speech, and Signal Processing, March 2016.

Related publications

E. Benetos, G. Lafay, M. Lagrange, and M. Plumbley, “Detection of overlapping acoustic events using a temporally-constrained probabilistic model,” in IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2016, pp. 6450–6454.: [More Details] [BIBT_EX] [URL (ext.)]; @inproceedings{emmanouil2016a, author = {Emmanouil Benetos and Gregoire Lafay and Mathieu Lagrange and Mark Plumbley}, title = {Detection of overlapping acoustic events using a temporally-constrained probabilistic model}, booktitle = {IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)}, month = {March}, organization = {IEEE}, pages = {6450-6454}, year = {2016} }

Members

Manager: Emmanouil Benetos, Gregoire Lafay, Mark Plumbley, Mathieu Lagrange