arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.03514cs.SDcs.LGeess.AS
收藏

通过知识转移、数据增强和预训练最大化小数据集上的音频事件检测模型性能:一项消融研究

Maximizing Audio Event Detection Model Performance on Small Datasets Through Knowledge Transfer, Data Augmentation, And Pretraining: An Ablation Study

Microsoft(微软)

Daniel Tompkins, Kshitiz Kumar, Jian Wu

摘要:

Xception模型通过ImageNet权重的知识转移、AudioSet上的预训练和动态数据增强管道,在ESC-50数据集上达到最先进的(SOTA)音频事件检测精度。本文提出了一种消融研究,分析哪些组件有助于提高性能和训练时间。一个较小的Xception模型也提出了接近SOTA的性能,几乎三分之一的参数。

英文摘要:

An Xception model reaches state-of-the-art (SOTA) accuracy on the ESC-50 dataset for audio event detection through knowledge transfer from ImageNet weights, pretraining on AudioSet, and an on-the-fly data augmentation pipeline. This paper presents an ablation study that analyzes which components contribute to the boost in performance and training time. A smaller Xception model is also presented which nears SOTA performance with almost a third of the parameters.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑