Abir Messaoudi

Primary tabs

Biography

Abir Messaoudi is a Senior Research Engineer at the Language Technologies Unit of the Barcelona Supercomputing Center. She holds a Computer Engineering degree from the National Institute of Applied sciences and Technologies of Tunisia, a Research Master’s degree in Information Processing and Complexity of Living from The National School of Engineering of Tunis, and a Research Master's Degree in Mathematics and Computer Science from the University of Paris Cité. She works on innovative NLP and Speech projects prioritizing preserving and advancing under and mid-represented languages and dialects. 

Educació

 

  • Doctor of Philosophy - PhD, Spoken Language Understanding, Code-swiching ASR - University of Monastir, Tunisia (2020 - 2022)
  • Research Master's Degree in Mathematics and Computer Science - University of Paris Cité, France 2019.
  • Research Master's Degree in Information Processing and complexity of living - National School of Engineering of Tunis, Tunisia 2019.
  • Engineering Degree - National Institute of Applied Science and Technology, Tunisia 2019.

Research

  • 2024:
    • Messaoudi, Abir, et al. "ILENIA_VOZ ASR System Fusion for Albayzin 2024 Speech to Text Challenge." Proc. IberSPEECH 2024. 2024.
  • 2023:
    • Haddad, H., Rouhou, A. C., Messaoudi, A., Korched, A., Fourati, C., Sellami, A., ... & Ghriss, F. (2023). TunBERT: pretraining BERT for Tunisian dialect understanding. SN Computer Science.
  • 2022:
    • A. Messaoudi, et al. "TuniSER ‑ Toward a Tunisian Speech Emotion Recognition System." International Conference on Natural Language and Speech Processing (ICNLSP)
    • A. Messaoudi, et al. "TunBERT Pretrained Contextualized Text Representation for Tunisian Dialect."  International Conference on Intelligent Systems and Patterns Recognition (ISPR)
  • 2021:
    • A. Messaoudi, et al. "Tunisian Dialectal End‑to‑end Speech Recognition based on DeepSpeech.", Elsevier, Procedia Computer Science, International Conference on AI in Computational Linguistics.
    • C. Fourati, H. Haddad, A. Messaoudi, M. BenHajhmida, A. Ben Elhaj Mabrouk, and M. Naski. "Introducing a large tunisian arabizi dialectal dataset for sentiment analysis". Arabic Natural Language Processing Workshop (WANLP).
    • Mabrouk, A. B. E., Hmida, M. B. H., Fourati, C., Haddad, H., & Messaoudi, A. "A Multilingual African Embedding for FAQ Chatbots."
  • 2020:
    • A. Messaoudi , et al, "Sentiment Analysis in Tunisian Social Media Using Deep Learning." Swedish Language Technology Conference (SLTC).
    • A. Messaoudi, et al. "Learning Word Representations for Tunisian Sentiment Analysis." The Mediterranean Conference on Pattern Recognition and Artificial Intelligence.
    • C. Fourati, A. Messaoudi, H. Haddad. "TUNIZI: a Tunisian Arabizi sentiment analysis Dataset." AfricaNLP Workshop, Putting Africa on the NLP Map.
    • A. Messaoudi, et al. "iCompass at SemEval‑2020 Task 12: From a Syntax‑ignorant N‑gram Embeddings Model to a Deep Bidirectional Language Model."  Proceedings of the Fourteenth Workshop on Semantic Evaluation.