ZarahShibli/Arabic_Punctuation_Prediction

Sequence to sequence model for Arabic punctuation prediction.

19
/ 100
Experimental

This project helps anyone working with spoken Arabic text that lacks proper punctuation. It takes unpunctuated Arabic text, especially output from an Automatic Speech Recognition (ASR) system, and automatically adds common punctuation marks like commas, periods, semicolons, and question marks. This makes the text much clearer and easier to read and understand.

No commits in the last 6 months.

Use this if you regularly work with transcribed Arabic speech or raw Arabic text that is missing punctuation and needs to be made more readable.

Not ideal if your Arabic text already contains punctuation, or if you require prediction for a wider range of punctuation marks beyond the five common ones supported.

Arabic-text-processing speech-to-text-post-editing text-readability language-tools
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 6 / 25

How are scores calculated?

Stars

12

Forks

1

Language

Jupyter Notebook

License

Last pushed

Feb 13, 2020

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ZarahShibli/Arabic_Punctuation_Prediction"

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