Logo image
Back
Cross-Referencing Self-Training Network for Sound Event Detection in Audio Mixtures
Journal article   Open access   Peer reviewed

Cross-Referencing Self-Training Network for Sound Event Detection in Audio Mixtures

Sangwook Park, David K. Han, Mounya Elhilali and Sungchul Park
IEEE transactions on multimedia, pp 1-1
26 May 2022
DOI: https://doi.org/10.1109/TMM.2022.3178591
url
http://arxiv.org/abs/2105.13392View
SubmittedOpen Access (License Unspecified),  Open

Abstract

Data models Event detection Perturbation methods Predictive models pseudo label self-training semi-supervised learning Semisupervised learning Sound event detection Task analysis Training
Sound event detection is an important facet of audio tagging that aims to identify sounds of interest and define both the sound category and time boundaries for each sound event in a continuous recording. With advances in deep neural networks, there has been tremendous improvement in the performance of sound event detection systems, although at the expense of costly data collection and labeling efforts. In fact, current state-of-the-art methods employ supervised training methods that leverage large amounts of data samples and corresponding labels in order to facilitate identification of sound category and time stamps of events. As an alternative, the current study proposes a semi-supervised method for generating pseudo-labels from unsupervised data using a student-teacher scheme that balances self-training and cross-training. Additionally, this paper explores post-processing which extracts sound intervals from network prediction, for further improvement in sound event detection performance. The proposed approach is evaluated on sound event detection task for the DCASE2020 challenge. The results of these methods on both "validation" and "public evaluation" sets of DESED database show significant improvement compared to the state-of-the art systems in semi-supervised learning.

Metrics

13 Record Views
5 citations in Scopus

Details

Logo image