Serenity

Inspiration

A big problem in today's fast-paced society is mental health. The hectic nature of many people's lives leaves little room for journaling and reflection. However, journaling has great benefits for mental health. It's a method of expressing yourself emotionally that can reduce stress and pent up feelings and allow for self-reflection. The problem is, people might find it hard to find time for it. Through this lack of attention, we find ourselves having deep-rooted thoughts and problems that we hide, stash, or keep in the back of our minds. That's why we, the Windsor Lancers, decided to tackle this problem head-on with an AI powered conversational journaling app!

What does it do?

As you enter Serenity, you come across a login page that allows you to login to access your data. Once you are logged into Serenity, you are then able to get into a menu page with two features: Users can create a new journal entry that includes a conversational feature where you can talk to an AI that will have a conversation with you and help you reflect. The second feature is a display of the logs of the previous conversation, including conversation logs and their sentiment scores.

How we built it

We planned and brainstormed our initial ideas on a whiteboard. We decided to use Flutter for the front end and Flask in the backend. We used MongoDB Realm to store and retrieve our data. We used the ChatGPT API and the google cloud NLP API to analyze messages and generate relevant question as well as calculate the sentiment scores.

Challenges we ran into

During this project, we've gotta say was very challenging as it is a 24 hour hackathon, which is a lot less than the standard 36 hour hackathons. A big problem we faced was running the project on different computers, even when sharing the exact files through Github, since our IDEs did have a lot of custom stuff and secret keys. Another challenge we really ran into was being able to implement features into flutter. Most of all, we faced a whole plethora of cocoa pods library issues, some taking up to hours to get our IDE working again such as this: https://github.com/CocoaPods/CocoaPods/issues/12081. Though overall, through teamwork, redbulls and hard work, we really pushed through with a great project in 24 hours!

Accomplishments that we're proud of

Under 24 hour Hackathon
Using a mostly new tech stack
Fixing the amount of problems we faced with Cocoa Pods
The app works and the backend was deployed succesfully ##What's next for Serenity Even though, we were able to accomplish a great amount for Serenity, having only 24 hours really restricted the amount of feature we wanted to implement. Some of the features/improvements that we would really look forward to implement:
A variety of graphs to display our data
Emotion detector through speech
Typewriter effect for response
More advanced conversations & analysis algorithms