scholar.google.com › citations
This article adapted four recent semi-supervised learning methods to the task of audio tagging, with MixMatch and FM being the best method in most ...
Feb 17, 2021 · In this article, we adapted four recent SSL methods to the task of audio tagging. The first two methods, namely Deep Co-Training (DCT) and Mean ...
Feb 16, 2021 · In this article, we adapted four recent SSL methods to the task of audio tagging. The first two methods, namely Deep Co-Training (DCT) and Mean ...
Improving Deep-learning-based Semi-supervised Audio Tagging with Mixup. L ... Weakly supervised and semi-supervised deep learning for sound event detection.
Improving Deep-learning-based Semi-supervised Audio Tagging with Mixup · Urban Sound Tagging using Convolutional Neural Networks · Semi-Supervised Audio ...
Results have shown that by applying Mixup, decision boundaries between classes can be improved [14]. MM obtained these results with versions of the “Wide ResNet ...
Sep 19, 2022 · The results reported in this article will highlight the positive impact of mixup in almost all our experiments. The article contributions are ...
In this article, we adapted five recent SSL methods to the task of audio classification. The first two methods, namely Deep Co-Training (DCT) and Mean ...
Mar 8, 2023 · However, when combined with mixup, ER is frequently improved. This can be partly explained by the fact that audio augmentations are often ...
To tackle this difficulty, we propose three strategies. The first is multitask learning using noisy data. The second is semi-supervised learning (SSL) using ...