A Window-based Approach to Data Augmentation for Text Normalization

This repository will contain the code for the paper "A Window-based Approach to Data Augmentation for Text Normalization."

TODO:

1. Port code from Google Repository--focus on character-based approach.--done
2. Convert code to handle SOTA libraries.--done
3. Restructure code to handle multiple inputs.--done
4. Develop data for (plausibly) Chumash and Pomo data, consent permitting.
5. Revisit handling issues in sents_util.py.--likely done, to revisit
6. Generate requirements.txt.