Inspiration
- We think it's cool.
What it does
- Uses AI to analyze your music taste through Spotify and give you a personalized drink recipe
- Uses a robot to take that drink recipe and dispense it
How we built it
- Frontend is made in React with Tailwind, we have a queuing system for drink recipes
- We use Spotify's Web API to authenticate a user and pull metadata about their music tastes
- We use ChatGPT's API to convert different music taste info (e.g. top artists, top tracks, genres) into a drink recipe
- We store drink information and client secrets on the cloud with Google Firebase
- We use a Python script to pull drink info from an endpoint and communicate to an ESP32 via serial
- For the robot, we CADded and laser cut a stand with two platforms. An upper platform holds six different cups for drinks, all of which funnel into a cup at the bottom which will hold the mixed drink. Each cup has a straw for dispensing liquid, and we control whether liquid is poured by using six servo motors to bend the straws up/down.
Challenges we ran into
- Working with the WiFi module / library problems with the ESP32
- Getting parts/resources for hardware, most places were closed for the weekend / the hurricane
- Figuring out the design of the robotics system.
Accomplishments that we're proud of
- Systems design, decoupling software and hardware so we could develop in parallel
- Low cost for the robot, mostly made with scrap parts
What we learned
- How to effectively work on hardware hacks
What's next for Mixtape Mocktails
- Adding audiovisual experiences based on music taste
- More variety of mixers
- Ice
Log in or sign up for Devpost to join the conversation.