An Approach for Self-Training Audio Event Detectors Using Web Data

TL;DR

This paper presents a self-training approach for audio event detection that leverages web data, specifically unlabeled YouTube audio, to enhance model performance beyond limited annotated datasets.

Contribution

It introduces a semi-supervised method combining labeled datasets with web audio to improve AED models, addressing data scarcity and acoustic diversity challenges.

Findings

Improved AED performance after self-training with web data

Web audio introduces challenges like noise and variability

Method demonstrates potential for scalable audio event detection

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.