An Unsupervised Hybrid Approach for Online Detection of Sound Scene Changes in Broadcast Content
×
Cite This
Citation & Abstract
G. Sevkin, A. Craciun, and T. Bäckström, "An Unsupervised Hybrid Approach for Online Detection of Sound Scene Changes in Broadcast Content," Paper P1-2, (2017 June.). doi:
G. Sevkin, A. Craciun, and T. Bäckström, "An Unsupervised Hybrid Approach for Online Detection of Sound Scene Changes in Broadcast Content," Paper P1-2, (2017 June.). doi:
Abstract: In this paper we describe an online system for broadcast content, which can detect sound scene changes with high accuracy. The system is unsupervised and does not require prior information on the segment classes. A scene change probability score is computed for each frame of the signal using a hybrid approach combining a model-based (Gaussian Mixture Model) with a distance-based (Hotelling’s T2-Statistic) segmentation method. The mixture model parameters are adapted online using the previous frames of the signal. Experiments on real recordings show that we can achieve more than 85% correct segment change detection with only 16% false detections.
@article{sevkin2017an,
author={sevkin, gökhan and craciun, alexandra and bäckström, tom},
journal={journal of the audio engineering society},
title={an unsupervised hybrid approach for online detection of sound scene changes in broadcast content},
year={2017},
volume={},
number={},
pages={},
doi={},
month={june},}
@article{sevkin2017an,
author={sevkin, gökhan and craciun, alexandra and bäckström, tom},
journal={journal of the audio engineering society},
title={an unsupervised hybrid approach for online detection of sound scene changes in broadcast content},
year={2017},
volume={},
number={},
pages={},
doi={},
month={june},
abstract={in this paper we describe an online system for broadcast content, which can detect sound scene changes with high accuracy. the system is unsupervised and does not require prior information on the segment classes. a scene change probability score is computed for each frame of the signal using a hybrid approach combining a model-based (gaussian mixture model) with a distance-based (hotelling’s t2-statistic) segmentation method. the mixture model parameters are adapted online using the previous frames of the signal. experiments on real recordings show that we can achieve more than 85% correct segment change detection with only 16% false detections.},}
TY - paper
TI - An Unsupervised Hybrid Approach for Online Detection of Sound Scene Changes in Broadcast Content
SP -
EP -
AU - Sevkin, Gökhan
AU - Craciun, Alexandra
AU - Bäckström, Tom
PY - 2017
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - June 2017
TY - paper
TI - An Unsupervised Hybrid Approach for Online Detection of Sound Scene Changes in Broadcast Content
SP -
EP -
AU - Sevkin, Gökhan
AU - Craciun, Alexandra
AU - Bäckström, Tom
PY - 2017
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - June 2017
AB - In this paper we describe an online system for broadcast content, which can detect sound scene changes with high accuracy. The system is unsupervised and does not require prior information on the segment classes. A scene change probability score is computed for each frame of the signal using a hybrid approach combining a model-based (Gaussian Mixture Model) with a distance-based (Hotelling’s T2-Statistic) segmentation method. The mixture model parameters are adapted online using the previous frames of the signal. Experiments on real recordings show that we can achieve more than 85% correct segment change detection with only 16% false detections.
In this paper we describe an online system for broadcast content, which can detect sound scene changes with high accuracy. The system is unsupervised and does not require prior information on the segment classes. A scene change probability score is computed for each frame of the signal using a hybrid approach combining a model-based (Gaussian Mixture Model) with a distance-based (Hotelling’s T2-Statistic) segmentation method. The mixture model parameters are adapted online using the previous frames of the signal. Experiments on real recordings show that we can achieve more than 85% correct segment change detection with only 16% false detections.
Authors:
Sevkin, Gökhan; Craciun, Alexandra; Bäckström, Tom
Affiliations:
International Audio Laboratories, Erlangen, Friedrich- Alexander-Universität (FAU), Erlangen, Germany; Aalto University, Aalto, Finland(See document for exact affiliation information.)
AES Conference:
2017 AES International Conference on Semantic Audio (June 2017)
Paper Number:
P1-2
Publication Date:
June 13, 2017Import into BibTeX
Subject:
Semantic Audio
Permalink:
http://www.aes.org/e-lib/browse.cfm?elib=18765