TY - GEN
T1 - Sound Event Detection with Sequentially Labelled Data Based on Connectionist Temporal Classification and Unsupervised Clustering
AU - Hou, Yuanbo
AU - Kong, Qiuqiang
AU - Li, Shengchen
AU - Plumbley, Mark D.
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/5
Y1 - 2019/5
N2 - Sound event detection (SED) methods typically rely on either strongly labelled data or weakly labelled data. As an alternative, sequentially labelled data (SLD) was proposed. In SLD, the events and the order of events in audio clips are known, without knowing the occurrence time of events. This paper proposes a connectionist temporal classification (CTC) based SED system that uses SLD instead of strongly labelled data, with a novel unsupervised clustering stage. Experiments on 41 classes of sound events show that the proposed two-stage method trained on SLD achieves performance comparable to the previous state-of-the-art SED system trained on strongly labelled data, and is far better than another state-of-the-art SED system trained on weakly labelled data, which indicates the effectiveness of the proposed two-stage method trained on SLD without any onset/offset time of sound events.
AB - Sound event detection (SED) methods typically rely on either strongly labelled data or weakly labelled data. As an alternative, sequentially labelled data (SLD) was proposed. In SLD, the events and the order of events in audio clips are known, without knowing the occurrence time of events. This paper proposes a connectionist temporal classification (CTC) based SED system that uses SLD instead of strongly labelled data, with a novel unsupervised clustering stage. Experiments on 41 classes of sound events show that the proposed two-stage method trained on SLD achieves performance comparable to the previous state-of-the-art SED system trained on strongly labelled data, and is far better than another state-of-the-art SED system trained on weakly labelled data, which indicates the effectiveness of the proposed two-stage method trained on SLD without any onset/offset time of sound events.
KW - Sound event detection
KW - connectionist temporal classification
KW - convolutional recurrent neural network
KW - sequentially labelled data
KW - unsupervised clustering
UR - http://www.scopus.com/inward/record.url?scp=85068988032&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2019.8683627
DO - 10.1109/ICASSP.2019.8683627
M3 - Conference Proceeding
AN - SCOPUS:85068988032
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 46
EP - 50
BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
Y2 - 12 May 2019 through 17 May 2019
ER -