Inspiration
Our inspiration came from an interesting combination of ideas, ChatGPT and Dungeons and Dragons. We found that it can be hard to find someone with the storytelling abilities required to be a great dungeon master. We also wanted to make a fun game, but for it to be something that really fires the imagination rather than just functioning as a distraction.
What it does
Our mobile app is like a chatbot, except that, behind the scenes, it sends custom prompts to the OpenAI API that help guide the user along in a "choose your own adventure story". The prompts are carefully crafted to allow a user maximum freedom on their adventures. The user is given multiple options for how to continue at every turn, including the option to make up their own path forward. When the adventure ends, a new one begins!
How we built it
When we began, we essentially split into two teams. Each team had their own ideas of what tech stack to use and what design to pursue. The race was on to see who would create a viable, functional prototype first, the Android group, or the React Native group. The latter group took a bold approach, and it payed off! They finished first with a beautiful UI that would run on many different platforms. They had fetched an API key and wrote most of their code in Java and JS. After that, the other team worked on the React project as well and tried to transfer their previous work into the app.
Challenges we ran into
When we came together, we had two projects that were very hard to blend. We all had no experience with React and very little mobile experience, so blending two projects with different tech stacks proved too difficult. The OpenAI API was also not very responsive. At first we could not find out why it wouldn't respond (our free trial had expired.) Then, we just found that the responses came in slower than we'd like.
Accomplishments that we're proud of
We all learned a lot and we also made an end product that is actually worth downloading! Each of us wrote a lot of code and explored mobile development in a way we had never done before.
What we learned
We learned about how to utilize an API key. We learned the security etiquette of hiding such keys. We learned a bit about how OpenAI responds to different prompt requests. We all learned a ton about mobile development and how to push our own boundaries!
What's next for storyTailor
We have plans to add a voice recognition and a text to speech component to storyTailor. It will be like talking to a great storyteller who wants you to interact with their tales. We also would like to better integrate some previous work we did that sends more custom prompts to OpenAI, depending on the user's responses.
Log in or sign up for Devpost to join the conversation.