Skip to main content Skip to navigation

Towards an AI to Win Ghana's National Science and Maths Quiz

Project Overview

The NSMQ AI project is an innovative initiative designed to create an AI system capable of competing in Ghana's National Science and Maths Quiz (NSMQ), a prestigious live competition for senior secondary school students. This project seeks to address the prevalent shortage of qualified teachers in Sub-Saharan Africa by providing AI-driven personalized learning support to students. The AI will utilize advanced technologies, including speech-to-text, text-to-speech, and question-answering systems, to enhance its performance during the competition. As the project is still in progress, it is set to debut in October 2023, with multiple teams collaborating on various components of the system to ensure its effectiveness. By leveraging generative AI, the project aims to not only enhance educational outcomes for participants in the quiz but also to serve as a model for integrating AI into broader educational contexts, thereby improving access to quality education in regions with limited teaching resources.

Key Applications

NSMQ AI project

Context: Educational competition for senior secondary school students in Ghana

Implementation: Developing a web application that integrates speech-to-text, question answering, and text-to-speech functionalities to compete in the NSMQ.

Outcomes: Potential for enhancing learning support for millions of students, addressing teacher shortages in Sub-Saharan Africa, and improving educational engagement through competition.

Challenges: Technical challenges in speech recognition for Ghanaian accents, ensuring real-time processing, and accurately answering questions.

Implementation Barriers

Technical Barrier

Challenges with finding affordable resources for model training and deployment, as well as speech recognition accuracy for Ghanaian accents. This includes difficulties in curating accurate datasets from past competitions due to scattered data sources and media coverage.

Proposed Solutions: Fine-tuning models on local datasets, collaborative efforts to gather relevant data, open-source contributions to improve the system, and using automation scripts to streamline data gathering and collaboration with teams for data annotation.

Project Team

George Boateng

Researcher

Jonathan Abrefah Mensah

Researcher

Kevin Takyi Yeboah

Researcher

William Edor

Researcher

Andrew Kojo Mensah-Onumah

Researcher

Naafi Dasana Ibrahim

Researcher

Nana Sam Yeboah

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: George Boateng, Jonathan Abrefah Mensah, Kevin Takyi Yeboah, William Edor, Andrew Kojo Mensah-Onumah, Naafi Dasana Ibrahim, Nana Sam Yeboah

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies