Final Demo
We have compiled a through demonstration of our full system. It has been split into two videos. The first is the AI Story Generator Application which can be found here. The second video walks through our mobile application which can be found here.
We have put together a slideshow to give viewers a broad overview of our product. You can view the slideshow here, or access a pdf of the slides here.
The development of the AI Generated Children’s Stories application was underpinned by thorough research and a keen understanding of both user needs and technological capabilities. This research-driven approach was pivotal in shaping the project's scope and technical architecture.
A significant portion of the research was devoted to understanding the target audience: parents, toddlers (4-6 years), and children (7-9 years). This involved demographic analysis, surveys, and interviews to gather insights into their technological proficiency, concerns about education, and preferences in interactive content. The findings directly influenced the application's design and functionalities to ensure they catered to the specific needs and interests of each user group.
Incorporating AI technologies was a cornerstone of this project. Research into natural language processing (NLP) and generative image models was conducted to select the most suitable AI frameworks. This led to the adoption of ChatGPT for text generation and MidJourney for image generation. These technologies were chosen for their proven capabilities in creating engaging and age-appropriate content.
The decision to operate the application as a free, non-profit initiative while leaving room for potential marketing collaborations with other companies was a result of extensive market research. This approach was chosen to maximize accessibility to reading materials for children, aligning with the overarching mission of the project.
The selection of the Layered Architecture and MVVM design patterns, as well as the choice of technologies (Flutter, Spring Boot, MySQL, etc.), was a result of careful consideration of scalability, maintainability, and performance requirements. These choices were validated through feasibility studies and prototyping to ensure they aligned with the project's objectives.
Research into potential risks, such as scope creep, resource constraints, technical difficulties, and communication breakdowns, led to the development of a robust risk management strategy. This included specific mitigation plans for each identified risk, ensuring that the project could adapt to unforeseen challenges.
The adoption of a Continuous Deployment with Monorepo and Gitflow model was informed by research into best practices for software development and deployment. This approach was chosen to facilitate seamless integration, testing, and deployment of updates, ensuring a consistent user experience.
In summary, the research conducted throughout the project lifecycle played a pivotal role in shaping the application's design, functionality, and overall strategy. This research-driven approach not only validated key decisions but also ensured that the final product was tailored to the specific needs and expectations of its intended users.
This information was gathered to complete our initial project planning document which can be viewed here as well as our System Requirements Specification document which can be viewed here.
