arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.12926eess.AScs.LGcs.SD
收藏

基于三元训练的半监督声事件检测

Semi-supervised Acoustic Event Detection based on tri-training

Toyota Technological Institute at Chicago(芝加哥丰田技术研究院);Amazon(亚马逊)

Bowen Shi, Ming Sun, Chieh-Chi Kao, Viktor Rozgic, Spyros Matsoukas, Chao Wang

摘要:

本文介绍了我们的工作,训练声事件检测(AED)模型使用未标记的数据集。最近的声学事件检测器基于大规模神经网络,这些神经网络通常使用大量标记数据进行训练。声学事件的标签获得起来很昂贵,并且相关的声学事件音频可能有限,特别是对于罕见事件。在本文中,我们利用互联网规模的未标记的数据集与潜在的域转移,以提高检测的声学事件。基于经典的三重训练方法,我们提出的方法在所有预定义的声学事件检测任务中,都显示了监督训练基线和半监督自训练设置的准确性提高。由于我们的方法依赖于集成模型,我们进一步表明,可以通过知识蒸馏将改进提炼为单个模型,从而使单个学生模型保持教师集成模型的高精度。

英文摘要:

This paper presents our work of training acoustic event detection (AED) models using unlabeled dataset. Recent acoustic event detectors are based on large-scale neural networks, which are typically trained with huge amounts of labeled data. Labels for acoustic events are expensive to obtain, and relevant acoustic event audios can be limited, especially for rare events. In this paper we leverage an Internet-scale unlabeled dataset with potential domain shift to improve the detection of acoustic events. Based on the classic tri-training approach, our proposed method shows accuracy improvement over both the supervised training baseline, and semisupervised self-training set-up, in all pre-defined acoustic event detection tasks. As our approach relies on ensemble models, we further show the improvements can be distilled to a single model via knowledge distillation, with the resulting single student model maintaining high accuracy of teacher ensemble models.

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。

↑