Sakayo Toadoum Sari

About

I am a Multilingual NLP researcher and engineer focusing on low-resource language technologies, including machine translation, ASR/TTS, and discourse-level semantics. Currently a PhD candidate at CRIL (Université d’Artois), I am passionate about responsible AI, dataset design, and inclusive language technologies for the Global South. I have worked with organizations such as UNICEF, EPFL, and Batazia, and actively contribute to Masakhane and AfricaNLP communities.

Skills

Resume

Download my latest CV here.

Projects

• Ngambay–French NMT: Developed low-resource MT model (AMMI Thesis).
• African ASR/TTS Models: Built and evaluated speech systems for African languages (CDIAL).
• Online Polarization Modeling: NLP-based research during internship at EPFL.
• UNICEF RCCE Mapping: Designed actor mapping & data collection tools in Chad.

Publications

• Pretraining Strategies using Monolingual and Parallel Data for Low-Resource MT. AfricaNLP 2025.
• AfriMTE & AfriCOMET: Empowering COMET for African Languages. arXiv 2023.
• MASAKHANEWS: News Topic Classification for African Languages. AfricaNLP @ ICLR 2023.
• Ngambay–French Neural Machine Translation. NLP4TIA @ RANLP 2023.
• Error-Correction DNNs for South African Wildlife-Vehicle Collision Data. MDPI 2022.

View All Publications

Contact