A student-led research group from FPT University, focused on Artificial Intelligence and academic exploration in Data Science.
SpeedyLabX was founded by a passionate group of undergraduate AI majors at FPT University. Operating under the mentorship of Ms. Minh Thư, our group aims to cultivate a research-driven mindset while building practical, impactful projects that bridge the gap between academic theory and real-world application.
Our core interests include:
- Machine Learning & Deep Learning: From classical algorithms to advanced neural networks.
- Data Science & Analytics: Extracting meaningful insights from complex data.
- Applied & Academic AI Research: Conducting structured research aimed at publication.
To explore, experiment, and contribute to the AI landscape through disciplined research and real-world applications — fostering both individual growth and collective innovation.
Here are some of the projects we are actively developing and publishing.
Proactive Air Quality Forecasting and Health Alert System for Melbourne 
A comprehensive system leveraging multimodal data and explainable AI (XAI) to forecast air quality and issue proactive health alerts for the city of Melbourne, Australia.
This project represents SpeedyLabX’s first accepted academic paper, officially accepted for publication at the Australasian Joint Conference on Artificial Intelligence (AJCAI 2025).
- Conference: AJCAI 2025 — Accepted Paper 🎉
- Tech Stack:
Python,TensorFlow,XGBoost,SHAP,DiCE - Repository: melbourne-air-quality-forecast
- Keywords: Multimodal Forecasting · Environmental AI · Explainability · Public Health
This milestone marks the beginning of SpeedyLabX’s research publication journey, demonstrating our commitment to high-quality, reproducible, and socially impactful AI research.
Enhancing Multimodal Speech Emotion Recognition via a Conformer–GAT Fusion Architecture with XAI
A multimodal speech emotion recognition framework integrating acoustic and linguistic features through a Conformer–Graph Attention fusion architecture with explainability (XAI).
Targeting submission to IJCAI 2026.
- Tech Stack:
PyTorch,Transformers,SHAP,WavLM,BERT - Status:
In Progress - Repository: ser-conformer-gat-xai
... more projects will be featured as they are developed.
We are actively expanding our internal research portfolio with the long-term goal of open-sourcing selected projects and contributing to the broader AI community.
SpeedyLabX is currently an internal research group.
External collaboration is limited to academic partnerships and university-supervised projects.
For inquiries, please reach out directly to a member of the team or our academic mentor.
“Fostering the next generation of AI researchers from FPT University.”