An approach for self-training audio event detectors using web data
Publication Type
Conference Proceeding Article
Publication Date
8-2017
Abstract
Audio Event Detection (AED) aims to recognize sounds within audio and video recordings. AED employs machine learning algorithms commonly trained and tested on annotated datasets. However, available datasets are limited in number of samples and hence it is difficult to model acoustic diversity. Therefore, we propose combining labeled audio from a dataset and unlabeled audio from the web to improve the sound models. The audio event detectors are trained on the labeled audio and ran on the unlabeled audio downloaded from YouTube. Whenever the detectors recognized any of the known sounds with high confidence, the unlabeled audio was use to re-train the detectors. The performance of the re-trained detectors is compared to the one from the original detectors using the annotated test set. Results showed an improvement of the AED, and uncovered challenges of using web audio from videos.
Discipline
Artificial Intelligence and Robotics
Research Areas
Data Science and Engineering
Publication
2017 25th European Signal Processing Conference (EUSIPCO)
Identifier
10.23919/EUSIPCO.2017.8081532
Publisher
IEEE
City or Country
Kos, Greece
Citation
ELIZALDE, Benjamin; SHAH, Ankit; DALMIA, Siddharth; LEE, Min Hun; BADLANI, Rohan; KUMAR, Anurag; RAJ, Bhiksha; and LANE, Ian.
An approach for self-training audio event detectors using web data. (2017). 2017 25th European Signal Processing Conference (EUSIPCO).
Available at: https://ink.library.smu.edu.sg/sis_research/6691
Additional URL
https://doi.org/10.23919/EUSIPCO.2017.8081532