期刊
APPLIED SCIENCES-BASEL
卷 11, 期 7, 页码 -出版社
MDPI
DOI: 10.3390/app11073043
关键词
deep learning; sound event detection; self-supervised learning; transfer learning; natural image
类别
资金
- Institute for Information & Communications Technology Promotion (IITP) - Korea government (MSIT) [2019-0-01335]
This study suggests that transfer learning from natural images can enhance performance in audio-related tasks, and self-supervised learning with natural images is an efficient pre-training scheme.
We propose the implementation of transfer learning from natural images to audio-based images using self-supervised learning schemes. Through self-supervised learning, convolutional neural networks (CNNs) can learn the general representation of natural images without labels. In this study, a convolutional neural network was pre-trained with natural images (ImageNet) via self-supervised learning; subsequently, it was fine-tuned on the target audio samples. Pre-training with the self-supervised learning scheme significantly improved the sound classification performance when validated on the following benchmarks: ESC-50, UrbanSound8k, and GTZAN. The network pre-trained via self-supervised learning achieved a similar level of accuracy as those pre-trained using a supervised method that require labels. Therefore, we demonstrated that transfer learning from natural images contributes to improvements in audio-related tasks, and self-supervised learning with natural images is adequate for pre-training scheme in terms of simplicity and effectiveness.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据