BYOL for Audio: Self-Supervised Learning for General-Purpose Audio Representation
Authors:
Top Cited Papers
- 18 July 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
Inspired by the recent progress in self-supervised learning for computer vision that generates supervision using data augmentations, we explore a new general-purpose audio representation learning approach. We propose learning general-purpose audio representation from a single audio segment without expecting relationships between different time segments of audio samples. To implement this principle, we introduce Bootstrap Your Own Latent (BYOL) for Audio (BYOL-A, pronounced “viola”), an audio self-supervised learning method based on BYOL for learning general-purpose audio representation. Unlike most previous audio self-supervised learning methods that rely on agreement of vicinity audio segments or disagreement of remote ones, BYOL-A creates contrasts in an augmented audio segment pair derived from a single audio segment. With a combination of normalization and augmentation techniques, BYOL-A achieves state-of-the-art results in various downstream tasks. Extensive ablation studies also clarified the contribution of each component and their combinations.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Contrastive Representation Learning: A Framework and ReviewIEEE Access, 2020
- Momentum Contrast for Unsupervised Visual Representation Learning2020
- Multi-Task Self-Supervised Learning for Robust Speech Recognition2020
- Optuna2019
- Look, Listen, and Learn More: Design Choices for Deep Audio Embeddings2019
- Unsupervised Learning of Semantic Audio Representations2018
- Deep Residual Learning for Image Recognition2016
- A Dataset and Taxonomy for Urban Sound Research2014
- Optimization Models2014
- DAFX: Digital Audio Effects2011