YerevaNN/translit-rnn

Automatic transliteration with LSTM

36
/ 100
Emerging

This tool helps language professionals, researchers, or anyone working with text to standardize inconsistently romanized text back into its original script. You provide a dataset of text in a specific language, including examples of its romanized form, and the tool will convert new romanized text into its proper script. It's designed for users dealing with languages that have varying romanization conventions, initially proven for Armenian.

No commits in the last 6 months.

Use this if you need to consistently convert text that has been written using Latin characters (romanized) back into its original, non-Latin script, especially when dealing with inconsistent romanization styles.

Not ideal if you need to convert text from its native script into a romanized form, or if you require support for a language not easily configured with character-level transliteration rules.

linguistics text-processing language-standardization digital-humanities multilingual-data
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 19 / 25

How are scores calculated?

Stars

93

Forks

20

Language

Python

License

Last pushed

Dec 07, 2018

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/YerevaNN/translit-rnn"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.