SayCheese.ai is a modern, fast, and highly secure web application built with React and Python that revolutionizes how educational institutions track attendance. By utilizing advanced machine learning pipelinesβincluding both Facial Recognition and Voice BiometricsβSayCheese.ai eliminates manual roll-calls and buddy punching, providing a seamless experience for both teachers and students.
- Create & Manage Subjects: Easily set up subjects and sections.
- AI-Powered Attendance:
- πΈ Face Recognition: Capture the classroom and automatically identify present students.
- π£οΈ Voice Recognition: Mark attendance seamlessly via student voice recordings.
- Generate QR Codes: Instantly create QR codes for students to enroll in classes.
- View Insights: See detailed attendance logs and track student enrollment.
- Simple Enrollment: Join a subject instantly using a unique join code or by scanning a QR Code.
- Track Progress: View attendance records for all enrolled subjects.
- Biometric Profiles: Register Face and Voice embeddings securely for automated check-ins.
The core of SayCheese.ai operates on an efficient ML pipeline backed by a scalable Supabase database.
graph TD
subgraph UI [React Frontend]
H[Home Screen]
T[Teacher Dashboard]
S[Student Dashboard]
end
subgraph ML [AI Pipelines]
FR[Face Pipeline<br/>dlib & SVM]
VR[Voice Pipeline<br/>Resemblyzer]
end
subgraph DB [Supabase Backend]
Teachers[(Teachers Table)]
Students[(Students Table)]
Subjects[(Subjects Table)]
Attendance[(Attendance Logs)]
end
H --> T & S
T -->|Creates Class/QR| Subjects
S -->|Enrolls via Code| Subjects
T -->|Classroom Image| FR
T -->|Audio Snippet| VR
FR -->|Identifies IDs| Attendance
VR -->|Identifies IDs| Attendance
FR -.->|Reads Embeddings| Students
VR -.->|Reads Embeddings| Students
- Frontend: React, Vite, React Bootstrap
- Backend/Database: Python, Supabase, bcrypt (Auth)
- Machine Learning Pipelines:
dlib&face_recognition_models: High-accuracy face detection and 128D encodings.scikit-learn(SVC): Linear SVM classifier for fast facial matching.resemblyzer&librosa: Deep learning based voice embedding extraction and audio processing.
- Utilities:
numpy,pandas,segno(QR code generation),pillow
Follow these steps to set up the project locally. The project is split into independent frontend and backend directories.
git clone https://github.com/yourusername/SayCheese.ai-app.git
cd SayCheese.ai-appNavigate to the frontend directory and start the development server:
cd frontend
npm install
npm run devNavigate to the backend directory and set up the virtual environment:
cd backend
python -m venv venv
# On Windows:
venv\Scripts\activate
# On Mac/Linux:
source venv/bin/activateInstall backend dependencies (make sure you have CMake installed for dlib):
pip install -r requirements.txtCreate an environment variables file in the backend directory and add your Supabase credentials:
[supabase]
url = "YOUR_SUPABASE_PROJECT_URL"
key = "YOUR_SUPABASE_ANON_KEY"# Start your Python backend server
python app.py
