Skip to content

Latest commit

 

History

History
36 lines (23 loc) · 556 Bytes

README.md

File metadata and controls

36 lines (23 loc) · 556 Bytes

MMTL

Dataset

  • IEMOCAP

Requirements

  • torch=1.0.1
  • python=3.6
  • numpy
  • matplotlib
  • sklearn
  • codecs
  • librosa
  • threading
  • pydub

Data Processing

  • use meta_data_pro.py --args value

Model Training

  • use python reptile_meta_train.py --args value for meta-train stage
  • use python reptile_meta_test.py --args value for meta-test stage

Ploting Confusion Matrix

  • use plot_confmat.py to plot confusion matrix

Notes

Code for "Meta Multi-task Learning for Speech Emotion Recognition"

We will update our code gradually.