/Contextual-Transformer

Supplementary materials for the paper: CT-SAT: Contextual Transformer for Sequential Audio Tagging

For more details and samples, please see our homepage.


Code and model

If you are interested in the code of CT-SAT or want to train the model, please refer to the cTransformer (Contextual Transformer) here .

And another model: GCT (Gated Contextual Transformer) on the DCASE2018 dataset can be found here .

Citation

Please feel free to use the sequential label dataset and consider citing our paper as

@inproceedings{hou22_interspeech,
  author={Yuanbo Hou and Zhaoyi Liu and Bo Kang and Yun Wang and Dick Botteldooren},
  title={{CT-SAT: Contextual Transformer for Sequential Audio Tagging}},
  year=2022,
  booktitle={Proc. Interspeech 2022},
  pages={4147--4151},
  doi={10.21437/Interspeech.2022-196}
}