YAD: Leveraging T5 for Improved Automatic Diacritization of Yorùbá Text

Akindele Michael Olawole, Jesujoba O. Alabi, Aderonke Busayo Sakpere, David I. Adelani
12/28/2024
cs.CL

No official code found. Best implementation: ajesujoba/YAD (2 stars, last commit 2025-09-26).

Abstract

In this work, we present Yorùbá automatic diacritization (YAD) benchmark dataset for evaluating Yorùbá diacritization systems. In addition, we pre-train text-to-text transformer, T5 model for Yorùbá and showed that this model outperform several multilingually trained T5 models. Lastly, we showed that more data and larger models are better at diacritization for Yorùbá

View on arXivView PDF

Code Implementations(1)

ajesujoba/YADCommunity100%
21Apr 1, 20241 years ago

Similar Papers

Are LLMs Good Text Diacritizers? An Arabic and Yorùbá Case Study

82% match2025

Bridging Gaps in Natural Language Processing for Yorùbá: A Systematic Review of a Decade of Progress and Prospects

79% match2025

Arabic Text Diacritization Using Deep Neural Networks

79% match2 repos3552019

Y-NQ: English-Yorùbá Evaluation dataset for Open-Book Reading Comprehension and Text Generation

79% match2024

Yankari: A Monolingual Yoruba Dataset

77% match2024

CATT: Character-based Arabic Tashkeel Transformer

77% match1 repo802024

Cite this paper

@article{olawole2024yad,
  title  = {YAD: Leveraging T5 for Improved Automatic Diacritization of Yorùbá Text},
  author = {Akindele Michael Olawole and Jesujoba O. Alabi and Aderonke Busayo Sakpere and David I. Adelani},
  year   = {2024},
  eprint = {2412.20218},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2412.20218v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2412.20218.svg)](https://codeofpaper.com/paper/2412.20218)

Discussion