Heavily Augmented Sound Event Detection utilizing Weak Predictions

The performances of Sound Event Detection (SED) systems are greatly limited by the difficulty in generating large strongly labeled dataset. In this work, we used two main approaches to overcome the lack of strongly labeled data. First, we applied heavy data augmentation on input features. Data augme...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-07
Main Authors: Nam, Hyeonuk, Byeong-Yun Ko, Lee, Gyeong-Tae, Seong-Hu, Kim, Won-Ho, Jung, Sang-Min, Choi, Yong-Hwa, Park
Format: Article
Language:English
Subjects:
Data augmentation
Datasets
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The performances of Sound Event Detection (SED) systems are greatly limited by the difficulty in generating large strongly labeled dataset. In this work, we used two main approaches to overcome the lack of strongly labeled data. First, we applied heavy data augmentation on input features. Data augmentation methods used include not only conventional methods used in speech/audio domains but also our proposed method named FilterAugment. Second, we propose two methods to utilize weak predictions to enhance weakly supervised SED performance. As a result, we obtained the best PSDS1 of 0.4336 and best PSDS2 of 0.8161 on the DESED real validation dataset. This work is submitted to DCASE 2021 Task4 and is ranked on the 3rd place. Code availa-ble: https://github.com/frednam93/FilterAugSED.
ISSN:2331-8422