Multi-Scale Time-Frequency Attention for Acoustic Event Detection

Jingyang Zhang, Wenhao Ding, Jintao Kang, Liang He


Most attention-based methods only concentrate along the time axis, which is insufficient for Acoustic Event Detection (AED). Meanwhile, previous methods for AED rarely considered that target events possess distinct temporal and frequential scales. In this work, we propose a Multi-Scale Time-Frequency Attention (MTFA) module for AED. MTFA gathers information at multiple resolutions to generate a time-frequency attention mask which tells the model where to focus along both time and frequency axis. With MTFA, the model could capture the characteristics of target events with different scales. We demonstrate the proposed method on Task 2 of Detection and Classification of Acoustic Scenes and Events (DCASE) 2017 Challenge. Our method achieves competitive results on both development dataset and evaluation dataset.


 DOI: 10.21437/Interspeech.2019-1587

Cite as: Zhang, J., Ding, W., Kang, J., He, L. (2019) Multi-Scale Time-Frequency Attention for Acoustic Event Detection. Proc. Interspeech 2019, 3855-3859, DOI: 10.21437/Interspeech.2019-1587.


@inproceedings{Zhang2019,
  author={Jingyang Zhang and Wenhao Ding and Jintao Kang and Liang He},
  title={{Multi-Scale Time-Frequency Attention for Acoustic Event Detection}},
  year=2019,
  booktitle={Proc. Interspeech 2019},
  pages={3855--3859},
  doi={10.21437/Interspeech.2019-1587},
  url={http://dx.doi.org/10.21437/Interspeech.2019-1587}
}