论文标题
弱标记的声音事件检测的多分支学习
Multi-Branch Learning for Weakly-Labeled Sound Event Detection
论文作者
论文摘要
弱监督的SED中隐含了两个子任务：音频标记和事件边界检测。将多任务学习与SED结合的当前方法需要这两个子任务的注释。由于在弱监督的SED中只有用于音频标记的注释，因此我们设计了具有不同学习目的的多个分支，而不是追求多个任务。与多个任务类似，多个不同的学习目的也可以防止多个分支共享的共同特征，从适合任何一个学习目的。我们根据不同的米策略和不同的合并方法的组合来设计这些多种不同的学习目的。 Dcase 2018 Task 4数据集和Urban-SED数据集的实验都表明我们的方法可以实现竞争性能。
There are two sub-tasks implied in the weakly-supervised SED: audio tagging and event boundary detection. Current methods which combine multi-task learning with SED requires annotations both for these two sub-tasks. Since there are only annotations for audio tagging available in weakly-supervised SED, we design multiple branches with different learning purposes instead of pursuing multiple tasks. Similar to multiple tasks, multiple different learning purposes can also prevent the common feature which the multiple branches share from overfitting to any one of the learning purposes. We design these multiple different learning purposes based on combinations of different MIL strategies and different pooling methods. Experiments on the DCASE 2018 Task 4 dataset and the URBAN-SED dataset both show that our method achieves competitive performance.
