Mehedi Hasan Bijoy
I am currently working as a Research Assistant at AaltoASR research group and pursuing my MSc degree at Aalto University, specializing in Computer, Communication, and Information Sciences, with a primary focus on Speech and Language Technology. Prior to this, I earned my BSc degree in Computer Science and Engineering with Summa Cum Laude distinction from NSU in 2021. I worked as a lecturer at BUBT, as a research assistant at IAR-UIU, and as a teaching assistant followed by a lab instructor at NSU.
My ongoing research includes accented English speech recognition and multilingual LLM compression, focusing on Multi-Teacher Knowledge Distillation and Multimodal Fusion.
Email  / 
CV  / 
LinkedIn  / 
Google Scholar  / 
GitHub
|
|
(† indicates equal contributions)
Publications
|
|
A transformer-based spelling error correction framework for Bangla and resource scarce Indic languages
Mehedi Hasan Bijoy†,
Nahid Hossain†,
Salekul Islam,
Swakkhar Shatabda
Computer Speech & Language, Volume: 89, Page: 101703
Code /
Paper
|
|
Panini: A Transformer Based Grammatical Error Correction Method for Bangla
Nahid Hossain†,
Mehedi Hasan Bijoy†,
Salekul Islam,
Swakkhar Shatabda
Neural Computing and Applications, Volume: 36, Issue: 7, Pages: 3463-3477
Code /
Paper
|
|
Advancing Bangla Punctuation Restoration by a Monolingual Transformer-based Method and a Large-Scale Corpus
Mehedi Hasan Bijoy†,
Mir Fatema Afroz Faria†,
Mahbub E Sobhani,
Tanzid Ferdoush,
Swakkhar Shatabda
EMNLP 2023, Proceedings of the First Workshop on Bangla Language Processing (BLP-2023)
Code /
Paper
|
|
Towards Sustainable Agriculture: A Novel Approach for Rice Leaf Disease Detection using dCNN and Enhanced Dataset
Mehedi Hasan Bijoy,
Nirob Hasan,
Mithun Biswas,
Suvodeep Mazumdar,
Andrea Jimenez,
Faisal Ahmed,
Mirza Rasheduzzaman,
Sifat Momen
IEEE Access, Volume: 12, Pages: 34174 - 34191
Code /
Paper
|
|
Image Tagging by Fine-tuning Class Semantics Using Text Data from Web Scraping
Mehedi Hasan Bijoy†,
Nirob Hasan†,
Md. Tahrim Faroque Tushar,
Shafin Rahman
ICCIT, 2021   (Oral Presentation)
Paper /
Oral Presentation
|
|
imgclassifier
(github)
Current Version : '0.0.2'
It is a naive image classifier developed on top of PyTorch. One can train a model and evaluate the performance of the model on a custom dataset by simply calling the train function of this library.
|
|