Skip to content

Latest commit

 

History

History
18 lines (12 loc) · 558 Bytes

README.md

File metadata and controls

18 lines (12 loc) · 558 Bytes

VaeRL²

Source code and datasets of paper: "VaeRL²: A VAE based Framework for Rap Lyrics Representation Learning".

Experiments

Here are three folds corresponding to three experiments mentioned in the paper:

  • nextline: Nextline Prediction;
  • song_generate: Dope-Lyrics Generation;
  • classification: Rap Genre Classification.

The experiment of parameter discussion is not summed up into a folder. It can be done with folder nextline by just modifying the corresponding parameters.

Dependencies

  • Python 3.5
  • Tensorflow == 1.1.0
  • Keras == 2.0