A Wrinkle In Time

Inspiration

Do you have someone in your life that is lonely and has a hard time using a keyboard? We have the solution for them! It struck us that while technology connects the world, it often leaves behind the very people who need connection the most.

We asked ourselves if we could make connecting online as barrier free as possible That was the idea for A Wrinkle In Time. We wanted to build a digital space where the keyboard and mouse are optional, and your voice is the only tool you need to find connect and play.

What it does

A Wrinkle In Time is a fully voice-controlled social platform designed to for the motor impaired.

Zero-Touch Interface: users don't need to touch a keyboard. They can navigate entirely by speaking to our AI agent. The website has generative audio guidance to help users navigate the platform.

Playful Onboarding: Instead of filling out boring forms users can simply speak their interests aloud. Avatar creation is done the same way and can be personalized to reflect their interests.

Smart Matching: The system matches you with others online who speak your language and share your hobbies.

Voice-First Gaming: Users can play games together that can be controlled by voice commands.

Custom Music: The site uses ElevenLabs to generate personalized background music to match the users tastes.

How we built it

The core of our application is built on a React frontend, but the magic happens through the ElevenLabs ecosystem.

Voice Control: We utilized ElevenLabs Conversational AI to act as the user interface of our site. We built custom client-side tools that allow the agent to trigger UI actions based on the user's intent.

Audio Generation: We used the ElevenLabs Sound Effects & Music API to generate custom background music. We wrote a logic layer that translates a user's casual description (e.g., "I love old-school jazz") into complex prompting for the model.

Games: The multiplayer games were written in Javascript with the adaptive control inputs translated using ElevenLabs.

Challenges we ran into

The biggest hurdle was latency and audio comprehension. When a user gives a command the system needs to understand it, process the intent, and update the game state for both players in real time.

Another challenge was prompt engineering for music. "Classical music" is too vague for a generator; we had to write a translation layer to expand simple user preferences into rich, descriptive prompts that resulted in high-quality audio.

Accomplishments that we're proud of

We made a website where the entire user journey from login to game over accessible without a single keystroke or click using ElevenLabs. We also created the three co-operative games ourselves using Javascript.

What we learned

We learned to use that accessibility drives innovation. By forcing ourselves to remove standard input methods (keyboard/mouse), we had to come up with creative solutions that actually made the UI cleaner and more intuitive for everyone, not just seniors. We also gained a deep appreciation for the nuance of conversational design—teaching an AI to understand not just words, but intent and context in a social setting.