Skip to content

Latest commit

 

History

History
38 lines (35 loc) · 1.25 KB

README.md

File metadata and controls

38 lines (35 loc) · 1.25 KB

HGFM

Pytorch implementation for codes in HGFM : A Hierarchical Grained and Feature Model for Acoustic Emotion Recgnition (ICASSP 2020)

Paper

HGFM : A Hierarchical Grained and Feature Model for Acoustic Emotion Recgnition. You can find this paper in IEEE Xplore

Please cite our paper if you find our work useful for your research:

@inproceedings{xu2020hgfm,
  title={HGFM: A Hierarchical Grained and Feature Model for Acoustic Emotion Recognition},
  author={Xu, Yunfeng and Xu, Hua and Zou, Jiyun},
  booktitle={ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages={6499--6503},
  year={2020},
  organization={IEEE}}

Model

HGFM framework

requirements

Package Version
torch 1.4.0
python 3.8.3
tqdm 4.47.0
matplotlib 3.2.2
numpy 1.18.5

Usage

Download the dataset feature

  1. Baidu Yun Disk[code: 6rez]
  2. Google Drive

Run the Code

  • Clone this repo and run the exe.sh
git clone https://github.com/thuiar/MMSA  
cd HGFM
bash exe.sh