Character-based orthographic variants
Open
enhancement
feat / cli
training
- Dominant language
- Python
- Stars
- 33.9k
- Forks
- 4.7k
- Avg merge
- 3m
- Merged PRs (30d)
- 1
Description
## Feature description
Similar to the token-based orthographic variants, it would be useful to add data augmentation options for character-based orthographic variants. Examples are the Romanian variants discussed in #4736 and German ß.
Contributor guide
Assessment
This issue has not been assessed yet.