Easy English

Ong, Shen Hoi (2024) Easy English. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RST_Ong Shen Hoi_Fulltext.pdf Restricted to Registered users only Download (3MB)

Abstract

In today's global context, many individuals lack confidence in speaking English, especially if it's not their native language. This challenge is notably prevalent in Malaysia, where Malay serves as the primary medium of education across all levels unless students attend English-focused schools. To address this issue, Easy English emerges as a VR-based language learning platform tailored for English learners. Easy English introduces a VR English Conversation game featuring ChatGPT as diverse NPCs (non-playable characters). Therefore, this game allows the learners having a more immersive and engaging environment to learn the English conversation compared to the traditional learning method by simulating the different types of real-life scenarios. This innovative approach offers a more engaging experience compared to traditional methods by simulating real-life scenarios. The game incorporates various techniques, including Whisper API for speech-to-text, ChatGPT for grammar checking and response prompts, TTS models for text-to-speech, Oculus Lip-sync for facial animation, and Ready Player Me for avatar creation. After the game finishes development, the Content Validation form has been conducted with the raters with high educational standards yielding an S-CVI/Universal Agreement of 0.6667, falling short of a perfect score. Similarly, the System Usability Scale (SUS) also provided to some participants and the System Usability Scale (SUS) feedback indicated a score of 50.71, meeting an 'OKAY' standard. Though playable, the project didn't attain optimal satisfaction due to user perceptions. In conclusion, Easy English presents an innovative method for English learning, contingent upon access to VR devices. While the project is functional, enhancements are necessary to elevate its effectiveness and user satisfaction