Inspiration
We wanted to create a way for people to control their computers without needing their hands, whether for accessibility or convenience. The idea came from imagining a future where anyone can use technology effortlessly through natural input methods like gaze and speech.
What it does
Iris lets users control their computers using only their eyes and voice. By tracking eye movement and recognizing voice commands, it enables actions like clicking, scrolling, and typing without any physical input. Iris provides both independence for users with disabilities and convenience for anyone seeking hands-free control.
How we built it
Iris was built using Python and Vite.
We used:
- $\textbf{OpenCV}$ to capture and preprocess webcam video input.
- $\textbf{MediaPipe}$ for real-time facial and eye tracking to detect gaze direction and small actions such as winks or squints.
- $\textbf{Vosk}$ for fast, offline speech recognition that enables accurate voice commands without needing an internet connection.
These components work together to interpret both vision and voice, creating an intelligent hands-free computer control system.
Challenges we ran into
One major challenge was achieving accurate gaze tracking in various lighting conditions and camera qualities. We also had to minimize response time between user input and on-screen actions to make the experience smooth and natural.
Accomplishments that we're proud of
We successfully created a working prototype that integrates both eye tracking and speech recognition in real time. We are especially proud of how Iris can empower users who might not have access to traditional input methods.
What we learned
We learned how to combine computer vision and speech transcription effectively to enable natural user interaction. We also gained experience optimizing real-time AI systems and improving accessibility through thoughtful design.
What's next for Iris
Next, we plan to expand Iris into a full desktop application with gesture recognition. We also aim to integrate cross-platform support so users can experience Iris on any device.

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