²»Á¼Ñо¿Ëù

Research Area: Information and Communication Technologies Research Group
Status:  Complete | Led By: Researcher- and student-led

The Challenge

Having grown up with ADHD and Dysgraphia, Paul McAllister is keenly aware of both the frustrations and rewards of learning to read and write. His quest to make the experience more accessible and fun led him to write a series of storybooks about a lovable monster named Herman. At the same time, he founded Monster House Publishing with the mission of improving literacy rates in New Brunswick and beyond.

Each child can experience unique reading challenges, for example with pronunciation, word recognition and retention, or staying focused. Paul was fortunate that patient family members, teachers, and librarians helped him overcome his struggles. Not all children have access to such individualized support. Believing new technologies like machine learning and augmented reality could help bridge that gap, Paul set out to create a new learning tool to keep new readers more engaged and motivated.

The Solution

Herman Literacy is an evidence-based mobile application supporting early readers in developing essential literacy skills. By incorporating artificial intelligence (AI) and augmented reality (AR), it provides an enhanced reading experience with personalized feedback.

The application creates a virtual world where Herman leaves the page to interact with budding readers. Through real-time AI-driven conversations, the animated character provides active encouragement and guidance. The system was developed to recognize different vocabulary levels and emotional cues and adapt responses accordingly—crafting an interactive reading experience customized to the individual child’s needs.

Every stage of the software development process was informed by research to ensure the end-product would meet the needs of both educators and students. One example is in the incorporation of game-like elements into the education process (referred to as “gamification”). This strategy is particularly helpful for motivating learners who do not see the value in reading because it is not modeled or encouraged at home.

As of summer 2025, the beta-stage application supports kindergarten to grade 5 children learning to read and write in English.

²»Á¼Ñо¿Ëù the Collaboration

Monster House Publishing collaborated with NBCC’s Information and Communication Technologies Research Group (ICTRG), and Let’s Nurture Technologies to create Herman Literacy. The Literacy Coalition of New Brunswick provided valuable insights and guidance at every phase of the project to ensure the application aligned with early childhood literacy standards.

New Brunswick-based digital solutions company Let’s Nurture Technologies got the ball rolling by creating an initial demo with AI integration. NBCC researchers then helped enhance the prototype by advising on early learning principles, designing and suggesting gamification elements, and improving the user flow.

Practicum students from NBCC’s IT: Game Development program contributed by researching the capabilities of various large language models (LLMs) and speech-to-text libraries to evaluate their performance and suitability for the project. They then created working demos implementing selected models to allow the development team to make informed choices as they continued testing and refining the application.

Research Outcomes

The Herman Literacy application shows great promise for enhancing reading proficiency and digital literacy across diverse student populations. The AI-driven assessment tools offer scalable solutions for broader adoption in Canadian education.

With core functionalities developed, the research team is continuing to test and refine Herman Literacy to validate its effectiveness before releasing it for broader use in schools. The team is continuing to refine the AI-driven conversation, comprehension analysis and gamification, with longer-term plans to incorporate additional interactive features and multilingual support.

Project Funder

Initial funding for this project was provided by the National Research Council of Canada’s Industrial Research Assistance program (NRC-IRAP). The work was further supported by the New Brunswick Innovation Foundation (NBIF), which provides funding to enable collaboration between research organizations and small and medium-sized enterprises in New Brunswick.