Orator is an AI-driven platform designed to help individuals improve their public speaking skills through real-time feedback powered by speech analysis, gesture tracking, and EEG-based emotion detection. The platform enables users to record practice sessions, receive data-driven insights, and track their progress over time, turning nervous practice into confident performance.
- Speech-to-Text Analysis – Uses Google Cloud Speech-to-Text to compare what you say with your uploaded script. Detects filler words, off-topic moments, and skipped key points.
- Gesture Detection (YOLOv11) – Tracks body language in real time. Detects gestures such as covering your face, standing too still, pacing, or looking down.
- EEG Emotion Tracking (Muse S Headband) – Reads stress and focus levels to detect nervousness or distraction.
- Performance Report – Summarizes clarity, confidence, and presence after each session to help you improve.
- Real-time Feedback Loop – Immediate insights help users adjust speech and posture live.
- Frontend: React + TypeScript + TailwindCSS
- Backend: Flask (Python)
- AI Models:
- Google Cloud Speech-to-Text API
- YOLOv11 for gesture detection
- BrainFlow + Muse EEG for emotion tracking
- Clone the Repository
git clone https://github.com/EEGenius/orator.git
cd orator- Backend Setup
cd backend
python -m venv venv
source venv/bin/activate # or venv\Scripts\activate on Windows
pip install -r requirements.txtCreate a file named .env or gcp_key.json for your Google Cloud credentials.
Run the Flask server:
python main.py- Frontend Setup
cd frontend
npm install
npm run dev- The user records a presentation through the frontend recorder.
- Audio data is sent to the Flask backend, which calls Google Cloud Speech-to-Text for transcription.
- The system compares speech with the uploaded script to analyze clarity and focus.
- YOLOv11 processes the video feed to detect key gestures and posture issues.
- EEG data from Muse tracks confidence and stress levels.
- A feedback report summarizes strengths and areas for improvement.
Orator follows a freemium model:
Free Tier: Basic recording and feedback.
Premium Tier: Advanced analytics, EEG insights, and longer sessions.
Enterprise: Institutional access for universities and training programs.
At Team EEGenius, we believe confidence can be learned and measured. Our mission is to make high-quality communication training accessible, affordable, and powered by AI.
Austin Bao, Jion Choi, Aqib Amran, David Xia, Raphael Ho