Skip to content

Latest commit

 

History

History
16 lines (10 loc) · 909 Bytes

File metadata and controls

16 lines (10 loc) · 909 Bytes

Diacritization

Papers

Title Summary Date Publisher
Arabic Text Diacritization Using Deep Neural Networks The results show that the neural Shakkala system significantly outperforms traditional rule-based approaches and other closed-source tools with a Diacritic Error Rate (DER) of 2.88% compared with 13.78%, which the best DER for the non-neural approach(obtained by the Mishkal tool). 2019 Shakkala

Frameworks

Name Publisher Techs Comments
Shakkala Jordan University of Science and Technology Python, Tensorflow, keras

More resources coming soon stay tuned ! 🤩 You are welcome to contribute to this project ! 🙏