Bright Buddy

Dialogic Reading Response
Flow Chart/Tech Stack

Inspiration

The inspiration behind Bright Buddy came from a deep understanding of the critical role social-emotional skills play in children's mental health and overall success in life. We realised that while parents and teachers recognize the importance of these skills, they often lack engaging and practical tools to teach them effectively. We saw an opportunity to combine our team’s expertise in child psychology with technology to create an AI-driven app that transforms storytime into an interactive, emotionally enriching experience.

What it does

Bright Buddy provides children with a fun, engaging way to learn empathy, emotional regulation, and resilience, while giving parents insights into their child’s well-being. It is designed to bridge the gap between emotional learning and technology, ensuring children grow up with the tools they need to thrive both emotionally and mentally.

How we built it

We initially used figma and react to create the frontend and to add the tutoring capabilities by adding open ai’s gpt 4 to help with the interactive dialogic reading process. Dialogic reading is a process where children are actively engaged in the reading content by critically thinking and conversing about the book content with a reading mentor. We trained the LLM with CROWD(completion, recall, open-ended questions, Wh- questions, Distancing) prompting techniques so that the AI reading mentor can act as a professional reading guide to read with children.

Challenges we ran into

We had to pivot our plans based on the time limit. For example, we made a backend but were unable to to use it due to time constraints and focused more on the frontend and training our model. Due to the lack of backend, we had to hard-code some of the images.

Accomplishments that we're proud of

We were able to implement a frontend that looks similar to our Figma designs and train a model to respond using dialogic reading.

What we learned

We learned to reach team consensus and adapt to our given constraints, learning to focus on key features and make up for the implementations we didn’t have time for.

What's next for Bright Buddy

Looking forward, we would like to add text-to-speech and speech-to-text APIs for when the child uses Bright Buddy without a parent to read with. In addition, we want to add full children's books into our database and edit our web app to be more adaptive.