# Thaka at KSAA-2026 Task 2: Regularized Fine-Tuning for Arabic Speech Diacritization

arXiv: https://arxiv.org/abs/2605.25928 · HTML: https://mlsanity.com/paper/2605.25928/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 14 references that are in the map.
- First posted on arXiv 25 May 2026.

## Builds on (most cited first, 8 of 8)

- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Optuna: A Next-generation Hyperparameter Optimization Framework](https://mlsanity.com/paper/1907.10902/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [R-Drop: Regularized Dropout for Neural Networks](https://mlsanity.com/paper/2106.14448/) (2021)
- [CATT: Character-based Arabic Tashkeel Transformer](https://mlsanity.com/paper/2407.03236/) (2024)
- [Abjad AI at NADI 2025: CATT-Whisper: Multimodal Diacritic Restoration Using Text and Speech Representations](https://mlsanity.com/paper/2510.24247/) (2025)

## Abstract (arXiv)

We describe the winning system for Task 2 of the KSAA-2026 Shared Task on Arabic Speech Dictation with Automatic Diacritization. The task requires producing fully diacritized Arabic text from speech audio and undiacritized transcripts, with only 2,327 training samples available and no external data permitted. Our system fine-tunes CATT-Whisper, a character-level multimodal model combining a pretrained CATT text encoder with a frozen Whisper speech encoder. The key to our approach is training regularization: R-Drop consistency regularization, Optuna-optimized hyperparameters with high weight decay, and Focal Loss. At inference, we average 200 stochastic forward passes across four model checkpoints using Monte Carlo Dropout at the softmax probability level. The system achieves 23.26% WER on the primary leaderboard metric (with case endings, including no-diacritic positions), placing 1st among all participants.