Abdulhady Abas Abdullah, M.Sc.

Research Professional in Artificial Intelligence

Erbil, Kurdistan Region, Iraq

abdulhady.abas@ukh.edu.krd

+964 751-884-8928

Professional Summary

A motivated research professional with a Master's degree in Artificial Intelligence and over three years of research experience in speech, language, and multimodal processing. Currently a Research Assistant in AI, I co-developed the first end-to-end Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) systems for Kurdish. Author or co-author of 20+ research papers in areas spanning deep learning, natural language processing, and Large Language Models (LLMs).* My current research focuses on LLM optimization, prompt engineering, vision-language models (VLMs), Audio LLMs, and multimodal AI systems.

20+

Publications

Years Experience

Awards Won

170+

Hours Speech Corpus

101

Total Citations

h-index

i10-index

Research Interests & Focus

Multilingual NLP & LLMs

Optimizing large language models (LLMs) for low-resource and multilingual applications.

Deep Learning & Transformers

Efficient architectures, fine-tuning, and prompt engineering for NLP tasks.

Speech & Multimodal AI

End-to-end ASR, Audio LLMs, and vision-language models (VLMs).

Education

2020–2022

M.Sc. in Artificial Intelligence

Soran University – Erbil, Iraq

Thesis: "Kurdish Speech Recognition using Deep Learning"

2016–2019

B.Sc. in Computer Science – Hardware

Soran University – Erbil, Iraq

Senior Project: "Design and Implementation of a Real Estate Web and Mobile Application"

Work Experience

2024–Present

Research Assistant in Artificial Intelligence

Artificial Intelligence and Innovation Centre (AIIC), University of Kurdistan Hewlêr (UKH), Erbil, Iraq

Conducting cutting-edge research in AI, specializing in speech processing, multilingual NLP, and Large Language Models (LLMs). Developed the first end-to-end Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) systems for Kurdish, bridging gaps in low-resource language technology. Research focuses on LLM optimization, prompt engineering, and adapting transformer models for Kurdish and other under-resourced languages. Collaborates with interdisciplinary teams to publish findings in high-impact conferences and journals.

2019–2022

Graduate Researcher (NLP/Speech Group)

Soran University & AsoSoft Research Team

Key R&D contributions in Kurdish language technologies:

Developed the first end-to-end ASR and TTS systems for Kurdish using deep learning
Created the first large-scale Kurdish speech corpus (170 hours) and text corpus (3M+ tokens)
Built the first large language model for Kurdish (CKB)
Developed OCR and handwriting recognition systems for Kurdish using modern deep learning architectures

Publications (Google Scholar)

Sentiment analysis in low-resource contexts: BERT’s impact on Central Kurdish

KM Awlla, H Veisi, Abdulhady Abas Abdullah

Language Resources and Evaluation, 2025 Published (Co-author, 3rd)

Machine Learning for Recruitment: Analyzing Job-Matching Algorithms

AK Jirjees, AM Ahmed, Abdulhady Abas Abdullah, J Lu, EM Noori, RN Kareem

Machine Learning 27, 1, 2025 Published (Co-author, 3rd)

In-depth Analysis on Machine Learning Approaches: Techniques, Applications, and Trends

Abdulhady Abas Abdullah, NS Mohammed, M Khanzadi, SM Asaad, ZK Abdul

ARO-THE SCIENTIFIC JOURNAL OF KOYA UNIVERSITY 13 (1), 190-202, 2025 Published (First author)

A novel facial recognition technique with focus on masked faces

Dana A. Abdullah, Dana R. Hamad, Ismail Y. Maolood, Hakem Beitollahi, Aso K. Ameen, Sirwan A. Aula, Abdulhady A. Abdullah, Mohammed Y. Shakorf, Sabat S. Muhamad

Ain Shams Engineering Journal 16 (5), 103350, 2025 Published (Co-author, 7th)

Do Optimization Algorithms Hold the Key to Advancing AI and Machine Learning? A Comprehensive Meta-Analysis Review

HM Ali, MY Hamza, TA Rashid, BA Hassan, Abdulhady Abas Abdullah

2025 Published (Co-author, 5th)

Backtracking Search Optimization Algorithm: A Practical Tutorial

BA Hassan, AA Hassan, AM Ahmed, TA Rashid, Abdulhady Abas Abdullah

Multi-objective Optimization Techniques, 119-132, 2025 Published (Co-author, 5th)

Kurdish end-to-end speech synthesis using deep neural networks

Sabat S. Muhamad, Hadi Veisi, Aso Mahmudi, Abdulhady Abas Abdullah, Farhad Rahimi

Natural Language Processing Journal 8, 100096, 2024 Published (Co-author, 4th)

Effective Facial Expression Recognition System Using Artificial Intelligence Technique

IS Yousif, TA Rashid, AS Shamsaldin, SA Abdulhameed, Abdulhady Abas Abdullah

Kurdistan Journal of Applied Research 9 (2), 117-130, 2024 Published (Co-author, 5th)

From Dialect Gaps to Identity Maps: Tackling Variability in Speaker Verification

Abdulhady Abas Abdullah, S Badawi, DA Abdullah, DR Hamad, HA Taher

arXiv preprint arXiv:2505.04629, 2025 Preprint (First author)

Video Forgery Detection for Surveillance Cameras: A Review

NB Tayfor, TA Rashid, SM Qader, BA Hassan, MH Abdalla, J Majidpour, Abdulhady Abas Abdullah

arXiv preprint arXiv:2505.03832, 2025 Preprint (Co-author, 7th)

Speaker Diarization for Low-Resource Languages Through Wav2vec Fine-Tuning

Abdulhady Abas Abdullah, SHT Karim, SA Ahmed, KR Tariq, TA Rashid

arXiv preprint arXiv:2504.18582, 2025 Preprint (First author)

A proposed chaotic dynamic cat swarm for the travelling salesman problem

AM Ahmed, BA Hassan, AS Qosaeri, Abdulhady Abas Abdullah, TA Rashid

Authorea Preprints, 2025 Preprint (Co-author, 4th)

Effect of Information Technology on Job Creation to Support Economic: Case Studies of Graduates in Universities (2023-2024) of the KRG of Iraq

AK Bapir, IY Maolood, DA Abdullah, AK Ameen, Abdulhady Abas Abdullah

arXiv preprint arXiv:2501.04438, 2025 Preprint (Co-author, 5th)

Breaking walls: Pioneering automatic speech recognition for Central Kurdish: End-to-end transformer paradigm

Abdulhady Abas Abdullah, H Veisi, T Rashid

arXiv preprint arXiv:2406.02561, 2024 Preprint (First author)

Optimizing feature selection with genetic algorithms: a review of methods and applications

ZY Taha, Abdulhady Abas Abdullah, TA Rashid

arXiv preprint arXiv:2409.14563, 2024 Preprint (Co-author, 2nd)

End-to-End Transformer-based Automatic Speech Recognition for Northern Kurdish: A Pioneering Approach

Abdulhady Abas Abdullah, S Tabibian, H Veisi, A Mahmudi, T Rashid

arXiv preprint arXiv:2410.16330, 2024 Preprint (First author)

NER-RoBERTa: Fine-Tuning RoBERTa for Named Entity Recognition (NER) within low-resource languages

Abdulhady Abas Abdullah, SH Abdulla, DM Toufiq, HS Maghdid, TA Rashid, PF Farho

arXiv preprint arXiv:2412.15252, 2024 Preprint (First author)

Advanced clustering techniques for speech signal enhancement: A review and metanalysis of fuzzy c-means, k-means, and kernel fuzzy c-means methods

Abdulhady Abas Abdullah, AM Ahmed, T Rashid, H Veisi, YH Rassul, B Hassan

arXiv preprint arXiv:2409.19448, 2024 Preprint (First author)

Enhancing Kurdish Text-to-Speech with Native Corpus Training: A High-Quality WaveGlow Vocoder Approach

Abdulhady Abas Abdullah, SS Muhamad, H Veisi

arXiv preprint arXiv:2409.13734, 2024 Preprint (First author)

Elevating Information System Performance: A Deep Dive into Quality Metrics

DA Abdullah, HA Khidir, IY Maolood, AK Ameen, DR Hamad, HS Beitolahi, Abdulhady Abas Abdullah

arXiv preprint arXiv:2412.18512, 2024 Preprint (Co-author, 7th)

Unveiling Public Service and Prosocial Motivation Through Machine Learning Insights

J Kabber, Abdulhady Abas Abdullah, AM Ahmed, T Rashid

2025 International Conference on Computer Science and Software Engineering, 2025 Preprint (Co-author, 2nd)

Central Kurdish automatic speech recognition using deep learning

Abdulhady Abas Abdullah, Hadi Veisi

Journal of the University of Anbar for Pure Science 16 (2), 108, 2022 Published (First author)

LLaMA-Adapter+ MRP: Integrating Meta-Reasoning Prompting with LLaMA-Adapter for Efficient Multi-Modal and Task-Adaptive Reasoning