Inspiration
We found a big gap in the current world of texting. We've often felt that people can sound "different" online and in-person, and we want to bridge that gap without demanding as big a time commitment as a video call. So, we use AI to detect our users' emotions when they read a message and add it to the chat.
What it does
The current app is a minimalistic chat web app. When somebody messages you and you open their message, we capture your emotions through the webcam. If the reaction was "significant", we add it to the chat.
How we built it
We built the frontend using ReactJS and the backend using Go. We used Redis as our database and used Socket.IO for real-time updates and syncing. The backend is hosted on an AWS EC2 instance. We're using AWS S3 to store our audio, and potentially other files (in the future)
Challenges we ran into
We spent a lot of time figuring out the Socket.IO connection and making the app reactive. We also tried getting CloudFlare to work which took a while.
Accomplishments that we're proud of
What we learned
What's next for Realer.app
We plan to expand Realer into a full-fledged chat application. The first feature we have in mind (that we've almost completed already) is adding reactions for audio messages, complete with timestamps of each reaction. This can then be expanded to a video as well. Further, we plan on leveraging Computer Vision to track eye movement and detect which part of the text message a reaction was associated with. The end-goal, potentially, is making Realer a mainstream chat application.
Log in or sign up for Devpost to join the conversation.