RCT: Random Consistency Training for Semi-supervised Sound Event Detection

2021/10/21 by Nian Shao, Shao, Nian, Erfan Loweimi +3
Computer Science · #Music and Audio Processing #Speech and Audio Processing #Speech Recognition and Synthesis

paper · pdf · doi:10.48550/arxiv.2110.11144

Abstract

Sound event detection (SED), as a core module of acoustic environmental analysis, suffers from the problem of data deficiency. The integration of semi-supervised learning (SSL) largely mitigates such problem while bringing no extra annotation budget. This paper researches on several core modules of SSL, and introduces a random consistency training (RCT) strategy. First, a self-consistency loss is proposed to fuse with the teacher-student model to stabilize the training. Second, a hard mixup data augmentation is proposed to account for the additive property of sounds. Third, a random augmentation scheme is applied to flexibly combine different types of data augmentations. Experiments show that the proposed strategy outperform other widely-used strategies.

Related

• Semi-supervised Sound Event Detection using Random Augmentation and Consistency Regularization
• Domestic sound event detection by shift consistency mean-teacher training and adversarial domain adaptation
• Sound Event Detection Based on Curriculum Learning Considering Learning Difficulty of Events
• HODGEPODGE: Sound event detection based on ensemble of semi-supervised learning methods
• SP-SEDT: Self-supervised Pre-training for Sound Event Detection Transformer