Inspiration
Shifty eyes, eyes too close together, a hand covering the face, drinking too much water too quickly - these are, as we know, foolproof forms of evidence for incrimination. However, behind computer screens, they no longer apply. In the game, Among Us, it can be very difficult to tell who is really on your side. We can't help you with that. (It's never good to make such broad generalizations.) But we can help you analyse your moments of fear, and help you see how often they come. Such is the implementation you will see below.
What it does
This web application records audio input from a user's microphone and analyses it with respect to the moments where the user says 'sus.' Following this, a statistical report of all the moments where 'sus' was uttered is given to the user, so they have a history of the moments in which they were most fearful. Not sure entirely why, it does seem kinda sus.
How we built it
We built this application in two parts: The backend, which we made using Google's App Engine (Python, Flask) to convert speech to text and analyse it, and a front end, which we made using ReactJS, NextJS and Vercel (the last for hosting). Files are communicated between the front and back ends through HTTP requests. Finally, special Among Us players were captured in sprite forms and locked into the application for eternity, ordered to move around for your amusement.
Challenges we ran into
Connecting the front and back ends, working with ReactJS, making the speech-to-text functionality fit for our purposes, and trying to finish in time.
Accomplishments that we're proud of
We made a fairly robust website, with a Web development system we weren't completely familiar with, and we worked hard to connect the front and back end.
What we learned
We learned how to delegate the work effectively in teams, use VSCode Live Share, use the programming languages and methods mentioned earlier, and hunt Among Us players in the wild.
What's next for 2022 Imposter Detector
To be bought by Jeff Bezos or Elon Musk. Apparently, in space, no one can hear you say 'sus'.
Built With
- flask
- google-app-engine
- nextjs
- python
- react
- speech-to-text
- vercel

Log in or sign up for Devpost to join the conversation.