About Mage Duel In the fantasy world of Degom, word magic, once the province of Mage Lords, is erupting across the land. Our story begins when a young commoner with emerging word magic is sent on a reconnaissance mission by her Mage Lord. She must gather intelligence about the Northern Warlord’s movements and actions all while navigating complex cultural environments full of diverse threats. Mage Duel is designed to accelerate language acquisition through bleeding-edge technologies, including the use of a natural language processing approach (NLP) grounded in sound language acquisition pedagogy. This is a serious game that supports language acquisition and translation competency from cradle-to-grave. Semantic Similarity Semantic similarity is how “close” two sentences are to each other based on meaning. The game currently includes one multilingual semantic similarity engine and is built for expansion. This state-of-the-art deep neural network-based model powers the translation learning activity. This means that in-game translations are evaluated based on meaning, which mirrors the way humans naturally acquire language skills. Thread Collection Thread Collection is a game mechanic implemented to develop phonological, morphological, and semantic discrimination skills. This activity Builds both receptive and productive vocabulary knowledge where the Learner must demonstrate mastery of targeted vocabulary words. It also includes audio-only activities. More Game Mechanics Mage Duel includes several other game mechanics such as: Timed language production activity Learning objectives of translation accuracy, fluency and complexity determine player performance rating and in-game behaviors Content is individualized based on the Mage Duel engine Will support multiple languages and language learning curricula in the future
